(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "The History of Ancient Iran (1983)"

Richard N. Frye 



THE HISTORY 

OF 

ANCIENT IRAN 









VerlagC-H-Beck 



THE HISTORY 

OF 
ANCIENT IRAN 



BY 

RICHARD N. FRYE 

AGA KHAN PROFESSOR OF IRANIAN 
HARVARD UNIVERSITY 




C. H. BECK'SCHE VERLAGSBUCHHANDLUNG 

MONCHEN 



HANDBUCH 
DER ALTERTUMSWISSENSCHAFT 

begrOndet von iwan VON MOLLER 

ERWEITERT VON WALTER OTTO 
FORTGEFOHRT VON HERMANN BENGTSON 

DRITTE ABTEILUNG, SIEBENTER TEIL 



Mit 3 Karten im Anhang 



CIP-Kurztitelaufnahme der Deutschen Bibliothek 

Handbuch der Alterlumswissenschaft / begr. von 
Iwan von Mliller. Erw. von Walter Otto. Fortgef. 
von Hermann Bengtson. - MUnchen : Beck 

NE: Mailer, Iwan von [Begr.]; Bengtson, 
Hermann [Hrsg.] 

Abt. 3. 

Teil 7. -» Frye, Richard N.: The history of ancient 

Iran 

Frye, Richard N.: 

The history of ancient Iran / by Richard N. Frye. - 

MUnchen : Beck, 1983. 

(Handbuch der Altertumswissenschaft : Abt. 3; 

Teil 7) 

ISBN 3 406 09397 3 



ISBN 3 406 09397 3 

C. H. Beck'sche Verlagsbuchhandlung (Oscar Beck) MUnchen 1984 
Satz: Spottiswoode Ballantyne Ltd., Colchester, England 
Druck: C. H. Beck'sche Buchdruckerei Nordlingen 
Printed in Germany 



Nels der Zukunft gewidmet 



TABLE OF CONTENTS 

Introduction IX 

Abbreviations XV 

Chapter I. Geographical Survey 1 

Chapter II. Demography 21 

Chapter III. Pre-Iranian History of the Plateau and Central Asia ... 45 

Chapter IV. Medes, Scythians and Eastern Rulers 65 

Chapter V. Achaemenids 87 

Chapter VI. Alexander the Great and the Seleucids 137 

Chapter VII. Greco-Bactrians, Sakas and Parthians 177 

Chapter VIII. The Parthians on the Plateau 205 

Chapter IX. The Kushans 249 

Chapter X. Minor Dynasties on the Plateau 271 

Chapter XI. The Sasanians 287 

Chapter XII. Eastern Iran and Central Asia 341 

APPENDICES 

Appendix 1 . Genealogical Tables of the Achaemenid, Parthian and Sasanian 

Dynasties 359 

Appendix 2. The Behistun Inscription of Darius near Kermanshah, Iran 363 

Appendix 3. The Bactrian Inscription of Surkh Kotal near Baghlan in 

Afghanistan 369 

Appendix 4. The Inscription of Shapur I at Naqsh-e Rustam in Fars . . 371 

Appendix 5. The MP Inscription of Narseh at Paikuli north of Khanaqin, Iraq 375 

Index 379 

Maps 



INTRODUCTION 



PROLEGOMENA 



The Leitmotiv of this book is the continuity in the history of western Iran from the 
earliest times to the Arab conquest, and in certain respects even down to the present. 
In spite of constant invasion and widespread destruction, like a phoenix the Persians 
have reasserted themselves often in the face of great adversity. The continuity is 
observable in many realms, one of the most striking of which is the language of 
ancient Persis, today the province of Fars, which developed from an Old Persian 
through a Middle to a New Persian stage with many admixtures, of course, and 
literary remains of all three stages exist. Of all the present day peoples of the Middle 
East, the Persians are the most conscious of their pre-Islamic past, and in many senses, 
traces of the ancient heritage of the area are better preserved in Persia than in Syria, 
Egypt, or elsewhere. 

In the present book the word 'Persia' refers to the contemporary land with the 
capital of Tehran. The term 'Iran,' however, is used for a greater, and unfortunately 
more imprecise, area including those regions where predominantly Iranian languages 
were spoken in antiquity. It must be emphasized that the term has no political 
connotations, and the word 'Iranian' is used primarily in a linguistic sense to 
differentiate the speakers of Iranian languages from Semitic speakers, Indians, or 
others. Persis or Parsa, on the other hand, is the modern province of Fars. 

The main source for Median and early Achaemenid history is the Ionian historian 
Herodotus, whose reliability, however, has been questioned many times. Many 
scholars, on the other hand, swear by his veracity when his data fits their theories, but 
reject or ignore him when his information goes counter to their reconstructions. On 
the whole, I believe one should follow the general rule that Herodotus generally 
reports what he heard and adds his own views in those matters where he moralizes, 
such as in the invasion of Greece by the Achaemenids, or the like. Whether his sources 
were reliable, of course, is another problem, but on the whole even if he reports stories 
which were current in his time, these stories are important for reconstructing ancient 
views. 

The second feature of this book, which is emphasized, is the extension of the Iranian 
cultural realm in antiquity to the borders of China, Siberia and South Russia. It is not 
the intention here to attempt to write a history of the steppes before the rise of the 
Turks but simply to stress the importance of Central Asia as an Iranian cultural area 
overshadowed by the Achaemenids and the Sasanians, but important in its own right 
as an independent source of Iranian ideas, art motifs, forms of society and government 
which had more of an impact on the Far East and Russia than heretofore realized. The 
post-Islamic-conquest flourishing of Bukhara, Samarqand and Khwarazm as great 
centers of Islamic learning and culture has puzzled many scholars who did not find the 
same phenomenon in the former Sasanian Empire. Central Asia was not a pale 
provincial reflection of the Sasanian Empire; rather it was a flourishing, independent 



X Introduction 

mercantile civilization before the coming of the Arabs. The profuse results of a 
plethora of Soviet archaeological excavations in Central Asia have not only 
enormously enriched our knowledge of this area in antiquity, but they have revised 
long held ideas that here nomads roamed and culture was almost non-existent. It is 
virtually impossible for one person to keep abreast of the ongoing work in Central 
Asia, both in the western and in the eastern (Sinkiang) regions. Although no attempt 
has been made to cover the history of Sinkiang or even adequately other parts of 
Central Asia, nonetheless constant reference to contacts between Central Asia and the 
more settled lands to the south, east and west has been made. It is hoped that the 
importance of this area, far from Europe, will be realized by the reader. The chapters 
of the present book pass from east to west and hopefully the differences as well as the 
affinities of the two Iranian areas will appear. Central Asia always has been bound 
with eastern Iran (present Afghanistan) rather than with Persia, and so it will be 
treated as part of eastern Iranian history. Continuity in the east was more frequently 
interrupted than in Persia, but it is also a feature of east Iranian culture. 

GENERAL WORKS ON THE ANCIENT HISTORY OF IRAN 

Unlike Greek, Roman, or even Egyptian and Mesopotamian history, there have been 
relatively few general histories of Iran down to the Arab conquests. The reasons for 
this lack are not difficult to discern. First, few historians have ventured into this area 
which requires formidable linguistic controls; instead the field has been left to 
philologists and linguists who are more interested in the detailed investigation of 
words than in the historical significance of texts. Second, the variety of aids one must 
utilize to reconstruct the history of Iran deters many investigators. One must turn to 
numismatics, art history, architecture, and, of course, archaeology to help recon- 
struct the past of a vast area, which again has repelled historians and encouraged 
archaeologists and art historians to take their place, often with not so happy results. 
Finally, so many travellers and amateurs have written about Iran, Afghanistan and 
Central Asia that scholars have shied away from writing a general work on the 
ancient history of the area, preferring instead to concentrate on much needed detailed 
monographs. It is true that much more investigation is needed on all aspects of the 
pre-Islamic history of this part of the world, but it is regrettable that syntheses of new 
discoveries in archaeology, art history, epigraphy and other branches of knowledge 
have not been made in a general history of the whole Iranian world before Islam. 
If we survey the books written on the general ancient history of Iran during the 
past one hundred years, the picture is quite different from that of other parts of the 
world which have received much more attention from scholars. The earliest general 
history of ancient Iran which utilized the native Old Persian inscriptions, primarily of 
Darius, was by George Rawlinson, brother of Henry, who copied and deciphered the 
Behistun inscription. 1 His first publication dealt with the Achaemenids, as the last of 
The Five Great Monarchies of the Ancient World, but this was soon augmented by the 

The two-volume history oflran, down to the reliable, while earlier books on ancient Iran by 
rise of the Sasanians, by le Comte de Gobineau. Thomas Hyde, Sylvestre de Sacy, among others, 
published in Paris in 1 869, is neither systematic nor did not pretend to be histories. 



Introduction XI 

Parthians and Sasanians to become The Seven Great Monarchies. 2 George Rawlinson, 
Professor of ancient history at Oxford, based most of his work on Classical sources, 
but he did use other sources available to him in translation and his work remained a 
standard history of ancient Iran for many years. It should be mentioned that the vast 
lands of eastern Iran (including Afghanistan) and Central Asia were largely 
untouched in Rawlinson's writings. This was rectified to some extent in a com- 
prehensive three- volume account of geography, ethnology, religions and history by 
Friedrich Spiegel, a philologist in the widest sense of the word, a professor of Oriental 
languages at the University of Erlangen over a century ago. In this encyclopedic 
work, Spiegel utilized Oriental as well as Classical sources. His historical survey was 
followed on the whole by Ferdinand Justi, who was Professor of Indogermanische 
Philologie at Marburg University, first in a general world history edited by W. 
Oncken in 1879, and then in the Grundrifi der iranischen Philologie, published in 
Strasbourg between 1896 and 1904. This latter publication long remained the 
standard survey of the ancient history of Iran. 

The popular work of Brigadier-General Sir Percy M. Sykes, A History of Persia, 
vol. 1 (London, 1915), contributed little, and the book of the Orientalist Clement 
Huart, La perse antique et la civilisation iranienne (Paris, 1925), is notable only for 
including results of the many years of excavation at the site of Susa, and the 
consequent addition of the history of the Elamites to the ancient history of Iran. The 
book by Huart was revised several times, and the title was changed to L'Iran antique, 
Elam et Perses, and remained the standard reference work in French until R. 
Ghirshman, an archaeologist, wrote his history called Iran (Payot, Paris, 1951), 
translated into English and published for the Penguin paperbacks in 1954. The work, 
as to be expected, was well illustrated and emphasis was on cultural history, art and 
archaeology. The same was true of the general history of ancient Iran by the Soviet 
Orientalist and archaeologist M. M. Dyakonov, published posthumously in 1961 . 3 In 
this work the eastern Iranian world, including Central Asia, first received more than 
cursory attention, primarily as a result of Soviet archaeological work there. This 
trend was further elaborated in the book of R. N. Frye, The Heritage of Persia 
(London, 1962), which, in spite of its title, chosen by the publisher for British readers, 
covers all areas where Iranian peoples lived in antiquity. The Cambridge History of 
Iran, vols. 2 and 3, will treat the early plus Achaemenid periods and the Parthians and 
Sasanians respectively, but even in those comprehensive volumes lacunae exist. It is 
clear that histories in the sense of a history of Greece, Rome, much less of France or 
England, cannot be written about ancient Iran; rather, a short sketch of ancient 
Iranian civilization, including art and archaeology as well as other fields, must be 
substituted in many periods. Nonetheless, an attempt is made here to utilize many 
works for a composite picture of the past, based on available sources. 

Naturally secondary sources have provided the basis for much of this com- 
prehensive history, even though disagreements may exist, since monographs and 
specialists on restricted periods and regions have been able to concentrate on problems 

2 His first work on the 'five monarchies' was 3 Ocherk istorii drevnego Iran (Moscow, 1961), 

published in London between 1 862 and 1 867, and 441 . 
the 'seven monarchies' came out in 1 873 and 1 875 
respectively. 



X Introduction 

mercantile civilization before the coming of the Arabs. The profuse results of a 
plethora of Soviet archaeological excavations in Central Asia have not only 
enormously enriched our knowledge of this area in antiquity, but they have revised 
long held ideas that here nomads roamed and culture was almost non-existent. It is 
virtually impossible for one person to keep abreast of the ongoing work in Central 
Asia, both in the western and in the eastern (Sinkiang) regions. Although no attempt 
has been made to cover the history of Sinkiang or even adequately other parts of 
Central Asia, nonetheless constant reference to contacts between Central Asia and the 
more settled lands to the south, east and west has been made. It is hoped that the 
importance of this area, far from Europe, will be realized by the reader. The chapters 
of the present book pass from east to west and hopefully the differences as well as the 
affinities of the two Iranian areas will appear. Central Asia always has been bound 
with eastern Iran (present Afghanistan) rather than with Persia, and so it will be 
treated as part of eastern Iranian history. Continuity in the east was more frequently 
interrupted than in Persia, but it is also a feature of east Iranian culture. 

GENERAL WORKS ON THE ANCIENT HISTORY OF IRAN 

Unlike Greek, Roman, or even Egyptian and Mesopotamian history, there have been 
relatively few general histories of Iran down to the Arab conquests. The reasons for 
this lack are not difficult to discern. First, few historians have ventured into this area 
which requires formidable linguistic controls; instead the field has been left to 
philologists and linguists who are more interested in the detailed investigation of 
words than in the historical significance of texts. Second, the variety of aids one must 
utilize to reconstruct the history of Iran deters many investigators. One must turn to 
numismatics, art history, architecture, and, of course, archaeology to help recon- 
struct the past of a vast area, which again has repelled historians and encouraged 
archaeologists and art historians to take their place, often with not so happy results. 
Finally, so many travellers and amateurs have written about Iran, Afghanistan and 
Central Asia that scholars have shied away from writing a general work on the 
ancient history of the area, preferring instead to concentrate on much needed detailed 
monographs. It is true that much more investigation is needed on all aspects of the 
pre-Islamic history of this part of the world, but it is regrettable that syntheses of new 
discoveries in archaeology, art history, epigraphy and other branches of knowledge 
have not been made in a general history of the whole Iranian world before Islam. 
If we survey the books written on the general ancient history of Iran during the 
past one hundred years, the picture is quite different from that of other parts of the 
world which have received much more attention from scholars. The earliest general 
history of ancient Iran which utilized the native Old Persian inscriptions, primarily of 
Darius, was by George Rawlinson, brother of Henry, who copied and deciphered the 
Behistun inscription. 1 His first publication dealt with the Achaemenids, as the last of 
The Five Great Monarchies of the Ancient World, but this was soon augmented by the 

The two-volume history of Iran, down to the reliable, while earlier books on ancient Iran by 
rise of the Sasanians, by le Comte de Gobineau, Thomas Hyde, Sylvestre de Sacy, among others, 
published in Paris in 1 869, is neither systematic nor did not pretend to be histories. 



Introduction XI 

Parthians and Sasanians to become The Seven Great Monarchies. 2 George Rawlinson, 
Professor of ancient history at Oxford, based most of his work on Classical sources, 
but he did use other sources available to him in translation and his work remained a 
standard history of ancient Iran for many years. It should be mentioned that the vast 
lands of eastern Iran (including Afghanistan) and Central Asia were largely 
untouched in Rawlinson's writings. This was rectified to some extent in a com- 
prehensive three- volume account of geography, ethnology, religions and history by 
Friedrich Spiegel, a philologist in the widest sense of the word, a professor of Oriental 
languages at the University of Erlangen over a century ago. In this encyclopedic 
work, Spiegel utilized Oriental as well as Classical sources. His historical survey was 
followed on the whole by Ferdinand Justi, who was Professor of Indogermanische 
Philologie at Marburg University, first in a general world history edited by W. 
Oncken in 1879, and then in the Grundrifi der iranischen Philologie, published in 
Strasbourg between 1896 and 1904. This latter publication long remained the 
standard survey of the ancient history of Iran. 

The popular work of Brigadier-General Sir Percy M. Sykes, A History of Persia, 
vol. 1 (London, 1915), contributed little, and the book of the Orientalist Clement 
Huart, La perse antique et la civilisation iranienne (Paris, 1925), is notable only for 
including results of the many years of excavation at the site of Susa, and the 
consequent addition of the history of the Elamites to the ancient history of Iran. The 
book by Huart was revised several times, and the title was changed to L'Iran antique, 
tilam et Perses, and remained the standard reference work in French until R. 
Ghirshman, an archaeologist, wrote his history called Iran (Payot, Paris, 1951), 
translated into English and published for the Penguin paperbacks in 1954. The work, 
as to be expected, was well illustrated and emphasis was on cultural history, art and 
archaeology. The same was true of the general history of ancient Iran by the Soviet 
Orientalist and archaeologist M. M. Dyakonov, published posthumously in 1 961 . 3 In 
this work the eastern Iranian world, including Central Asia, first received more than 
cursory attention, primarily as a result of Soviet archaeological work there. This 
trend was further elaborated in the book of R. N. Frye, The Heritage of Persia 
(London, 1962), which, in spite of its title, chosen by the publisher for British readers, 
covers all areas where Iranian peoples lived in antiquity. The Cambridge History of 
Iran, vols. 2 and 3, will treat the early plus Achaemenid periods and the Parthians and 
Sasanians respectively, but even in those comprehensive volumes lacunae exist. It is 
clear that histories in the sense of a history of Greece, Rome, much less of France or 
England, cannot be written about ancient Iran; rather, a short sketch of ancient 
Iranian civilization, including art and archaeology as well as other fields, must be 
substituted in many periods. Nonetheless, an attempt is made here to utilize many 
works for a composite picture of the past, based on available sources. 

Naturally secondary sources have provided the basis for much of this com- 
prehensive history, even though disagreements may exist, since monographs and 
specialists on restricted periods and regions have been able to concentrate on problems 

2 His first work on the 'five monarchies' was 3 Ocherk islorii drevnego Iran (Moscow, 1961), 

published in London between 1 862 and 1 867, and 441 . 
the 'seven monarchies' came out in 1873 and 1875 
respectively. 



XII Introduction 

which can only be touched by the present work, and reference is made to them for 
further information. For pre-Achaemenid history the writings of I. M. Dyakonov 
have proved especially valuable, while M. Dandamayev and W. Hinz have 
concentrated on the Achaemenids. Seleucid history in Iran is not so well served while 
J. Wolski is our man on the early Parthians as is V. G. Lukonin for the early Sasanians. 
Other scholars and their works are cited in the various chapters, but obviously it is 
impossible to cover or cite everything one should. Many times, rather than increase 
the volume by adding bibliographies, reference has been made to other works where 
extensive bibliographies are given. In the forthcoming age of computers and easily 
recalled bibliographies, the present book will become antiquated, but for the time 
being I hope it will be of some aid in finding one's way in the undergrowth and 
proliferation of writings on ancient Iran and Central Asia. 

TECHNICALITIES 

When one marvels at books of the past which were handset and printed with Greek 
and Oriental fonts by many experts, as well as a series of editors, one question which 
may arise is 'what price pedantry?' The use of diacritical points and macrons is 
necessary in linguistic arguments but hardly in a general history where those who 
read Oriental languages know what is meant, while to those who do not it is 
superfluous. Therefore I have omitted such banes to printers and proof-readers as 
much as possible, although in Arabic and other Semitic language words a more strict 
system of transliteration at times has been followed. I have tried to be consistent in 
spellings and only hope not too many slips will provide grist for reviewers. I must 
plead guilty to inconsistency in the usage of Latin and Greek forms of names and 
words, especially such details as final -us and -os. On the whole I have followed 
commonly accepted spellings except certain names, e.g. Priskos for Priscus, but again I 
hope that at least consistency has been observed in one and the same name. One 
explanation for variations, though not excuse, is that this book was originally 
intended to be written in German, but I found that my strength and ability by this 
time were not equal to the task and after much lost time and effort, I decided I would 
have to write in English. In the shift some 'un-English' forms of names may have 
survived. Also, in the case of transcripitions, rather than transliterations, of Middle 
Persian or other Iranian forms of names or words, consistency has not always been 
observed, for example: Weh not Veh but Nev rather than New, so one would not 
mistake the latter with the English word 'new.' Again, I hope no one will be confused ; 
specialists will not be, and those who are not should realize that there are many forms 
of transcription of foreign sounds and written forms. I have used the antiquated 
Wade-Giles system of transcribing Chinese characters, since it will be easy to consult 
the dictionaries and change into other forms of transcription with which I am not 
familiar. I must reiterate that this book is a general history of pre-Islamic Iran in its 
widest sense and not a philological treatise. 

The endings of Russian names cause trouble since forms in western European 
languages differ from the original. Thus we find Litvinskii written as Litvinsky and 
Perikhanyan written as Perikhanian, but hopefully the reader will understand such 
inconsistencies. 



Introduction XIII 

One innovation different from the usual interpretation of numismatics should be 
mentioned. Numismatists, on the basis of styles and iconography, assign coin series to 
certain time periods, and the suspicion that archaizing tendencies may negate their 
conclusions rarely becomes a factor to consider. Studies of Sogdian silver in the fifth 
and sixth centuries of our era have revealed archaic forms similar to similar objects of 
the Parthian and even Achaemenid periods. Coinage likewise may confuse one by 
archaic styles. Also the continuity of series of coins in eastern Iran or Central Asia, 
even one style lasting several centuries, as in Bukhara, frequently has been ignored by 
numismatists. In western Iran we find dynastic continuity while in the east we have 
lip-service to dynastic allegiance. Unfortunately, the history of eastern Iran is a vast 
area in itself and here it can only be touched with reference to bibliographies for 
further study. 

Finally, it is somewhat of a miracle that this volume appears since it was begun 
years ago when I was director of the Asia Institute of Pahlavi University in Shiraz, 
and I will not burden the reader with stories of the vicissitudes of lost books and 
portions of the manuscript, as well as the lack of aid in preparing the MS. This book 
should have been written by an old-fashioned Ordinarius with several assistants to 
compile bibliographies, fetch books, check references and read proofs. Unfortunately, 
I have been alone with many burdens and if it were not for the superb typing 
of Carolyn Cross, secretary of the Department of Near Eastern Languages and 
Civilizations at Harvard University, the book would not have seen the light of day. I 
would have preferred to have colleagues read the MS. and save me from errors, but 
today no one has time or inclination. Perhaps in the future machines will allow more 
time for such tasks but until then one should strive the best one can. If this work 
provides a useful guide to the history of ancient Iran I will be grateful. 

Richard N. Frye 

Hamburg, Shiraz and Cambridge, Mass. 



ABBREVIATIONS USED IN THIS BOOK 



Acta Antiqua Academiae Scientiarum Hungaricae (Budapest) 

Abhandlungen der Akademie in Wissenschaften und Literatur, Mainz 

Abhandlungen 

Altorientalische Forschung (Berlin) 

Abhandlungen der -» GWG 

Acta Iranica (Leiden) 

Annali dell' Istituto Orientale (Naples) 

American Journal of Archaeology (New York) 

American Journal of Ancient History (Cambridge, Mass.) 

American Journal of Semitic Languages (Chicago) 

Asia Major (London) 

Archaeologische Mitteilungen aus Iran (Berlin) 

American Numismatic Society (New York) 

American Numismatic Society Museum Notes (New York) 

Archiv Orientilnl (Prague) 

seeAbk.PAW 

Bulletin it I'icole francaise de I'Extrime Orient (Paris) 

Bibliotheca Orientalis (Amsterdam) 

Bulletin de la sociM linguistique (Paris) 

Bulletin of the School of Oriental (and African) Studies (London) 

Beitrage zur Namenforschung (Heidelberg) 

Central Asian Journal (Wiesbaden) 

Cambridge History of Iran (Cambridge, England) 

Centre national des recherches scientifiques (Paris) 

Comptes Renius de I'Acadimie des Inscriptions et Belles Lettres (Paris) 

Corpus Scriptorum Christianorum Orientalium (Paris) 

Darius, Naqsh-e Rustam inscription d in R. Kent, Old Persian (New Haven, 1953) 

Darius, Suez inscription 

Epigrqfika Vostoka (Leningrad) 

East and West (Rome) 

Fragmenta Historicorum Craecorum, ed. C. Mtlller (Paris) 

Die' Fragmente der Griechischen Historiker, ed. F. Jacoby (Berlin and Leiden) 

Gesellschaft der Wissenschaften zu Gdttingen 

Handbuch des Altpersischen, W. Brandenstein u. M. Mayrhofer (Wiesbaden, 1964) 

Handbuch der Orientalistik, ed. B. Spuler (Leiden) 

Iranica Antiqua (Leiden) 

lniogermanische Forschungen (Berlin) 

Indo-Iranian Journal (The Hague) 

Journal of the British Institute of Persian Studies (London) 

Istituto Italiano per il Medio ed Estremo Oriente (Rome) 

Journal Asiatique (Paris) 

Journal of the American Oriental Society (New Haven) 

Journal of the Economic and Social History of the Orient (Leiden) 

Journal of Indo European Studies (Washington, DC.) 

Journal of the K. R. Cama Oriental Institute (Bombay) 

Journal of the Oriental Society of Australia (Canberra) 

Journal of Near Eastern Studies (Chicago) 

Journal of the Numismatic Society of India (Benares) 

Journal of the Royal Asiatic Society (London) 

Kartir Ka'bah of Zardusht at Naqsh-e Rustam 

Kuhns Zeitschrift fur vergleichende Sprachwissenschaft (Gettingen) 



XVI 



Abbreviations 



MIA Materialy i Issledovaniya po Arkheologii SSSR (Moscow) 

MMAI Mtmoires de la mission archMogique en Iran (Paris) 

MDAFA Me" moires de la dtltgation archMogique franfaise en Afghanistan (Paris) 

MP Middle Persian (Pahlavi) 

MSS Miinchener Sludien zur Sprachwissenschaft (Munich) 

NC Numismatic Chronicle (London) 

NNM Numismatic Notes and Monographs (ANS) (New York) 

NP New Persian 

NTS Norsk Tidsskrift for Sprogvidenskap (Oslo) 

OAW Osterreichische Akademie der Wissenschaften (Vienna) 

OLZ Orientalische Literaturzeitung (Berlin) 

OP Old Persian 

OS Orientalia Suecana (Uppsala) 

PAW Preussische Akademie der Wissenschaften (Berlin) 

PO Patrologia Orientalis (Paris) 

PW Pauly Wissowa's RealenzyklopSdie der Altertumswissenschaft (Stuttgart) 

RIzA Revue des hudes Armtniennes (Paris) 

RHR Revue de I'histoire des religions (Pans) 

RN Revue numismatique (Paris) 

RSO Rivista degli Studi Orientali (Rome) 

SA Sovetskaya Arkheologiya (Moscow) 

Sb Sitzungsberichte 

SHAW Sitzungsber. der Heidelberger A.W. (Heidelberg) 

SWAW see Sb., WAW 

SI Studia Iranica (Paris) 

SKZ Shapur Ka'bah of Zardusht 

SPA Survey of Persian Art, ed. A. U. Pope (Oxford, 1939) 

TP Toung Pao (Leiden) 

TPS Transactions of the Philological Society (London) 

UCLA University of California in Los Angeles 

VDI Vestnik Drevnei Istorii (Moscow) 

WAW Wiener Akademie der Wissenschaften (Vienna) 

WZKM Wiener Zeitschrift fur die Kunde des Morgenlandes (Vienna) 

XPf Xerxes, Persepolis inscription f in R. Kent, Old Persian (New Haven, 1953) 

ZA Zeitschrift fur Assyriologie (Berlin) 

ZDMC Zeitschrift der Morgenldndischen Gesellschaft (Wiesbaden) 

ZII Zeitschrift fur Indologie und Iranistik (Leipzig) 



CHAPTER I 
GEOGRAPHICAL SURVEY 



The scholarly investigation of the geography of the Iranian plateau and Central Asia is 
relatively recent. Until the beginning of systematic collections of data on 
temperatures, water utilization, soils, and the like, geographical questions were mostly 
discussed by travellers and amateurs. Historical geography, on the other hand, was 
primarily the concern of academics who rarely if ever ventured into the field, and 
consequently misconceptions and gaps in knowledge abounded. Because of the 
strategic importance of Persia and Afghanistan, primarily British and Russian officers, 
but also others, sought to map the entire area at various times, and the result is today 
we have aerial photographs and maps of considerable detail. It is not possible to 
review all of the relevant geographical literature, but those works which aid the 
student of history will be mentioned under the three geographical headings of Iran, 
Afghanistan and Central Asia. 

Literature: Iran, general : The general gazetteer prepared by the army under the direction of Gen. Husain 
'All RazmarS' in ten volumes Farhang-e Joghriflya-ye Iran (Tehran, 1950-54), lists the towns and villages 
in each province, giving the location, population, main language, religion, principal occupations and 
industries, and similar information. Although in need of correction this collection of information is 
surpassed only by the comprehensive gazetteer of villages, based on the census of 1966, the Farhang-e 
Abadl-ha-ye Kishvar, of which 27 volumes have appeared in Tehran beginning in 1969. Another gazetteer 
in English based on the above, as well as on the British India Office archives is less comprehensive but 
useful. 1 Perhaps the most comprehensive textbook on the geography of Iran is the three-volume work of 
Mas 'od Kayhan,_/og/ird/iy<»-ye mufassal-e Iran (Tehran, 1 937), but now see E. Ehlers, Iran, Grundziige einer 
geographischen Landeskunde (Darmstadt, 1980). 

For the historical geography of Iran, we have F. Justi, Beitrage zur alten Geographie Persiens, Zwei 
Abtheilungen (Marburg, 1869), 56 S., who tried to identify place names in the Avesta with those of 
Classical and Islamic geographies. W. Tomaschek concentrated on Classical place names, especially in the 
Tabula Peutingeriana, a world map probably made by a Roman called Castorius in the third century but 
preserved in a thirteenth-century copy. He also identified many sites with place names in Arabic 
geographies. 2 The most erudite and productive historical geographer, however, was Joseph Marquart 
(later changed to Markwart) who clarified many identifications of site names in ancient Iran in his 
publications. 3 His commentaries to an edition and translation of an Armenian geography attributed to 
Moses of Chorene were perhaps his most valuable contribution to the historical geography of the entire 
Iranian world. 4 The old Geographie der Griechen und Rbmer of Konrad Mannert, fllnfter Theil (Leipzig, 
1829), is still useful. 

Sources: The classical books relevant to historical geography are first, Herodotus, with the useful Lexicon 
to Herodotus by J. E. Powell (Hildesheim, 1966), based on the edition of Hude (Oxford, 1926). Then we 
have Ptolemy in three editions by C. F. Nobbe (1 843 ; reprinted Hildesheim, 1 966), F. G. Wilberg (Essen, 
1845) and C. Muller (Paris, 1883), which is the basic work on place names in the pre-Christian east. The 

1 L. W. Adamec, ed.. Historical Gazetteer of Iran, von Iran, 2 Hefte (Gottingen, 1 896) ; "Beitrage zur 
1 (Graz, 1977) and foil. Geschichte und Sage von Eran," ZDMG, 49 

2 W. Tomaschek, "Zur historischen Topogra- (1895), 628-70, as well as his A Catalogue of the 
phievonPersien,"S6 WA W,\02 (1883), 146-231, Provincial Capitals of Eranshahr (Rome, 1931). 
and 108 (1885), 583-652, issued in book form in 4 Erinlahr nach der Geographie des Ps. Moses 
1972. Xorenac'i, Abh. GWG (Berlin, 1901). 



3 



Especially his Unlersuchungen zur Geschichte 



Chapter I 



Parthian Stations of Isidore of Charax, ed. by C. MUller in Geographi Graeci Minores (Paris, 1 855) and in 
Frg. Hist. (Dritter Teil C, Zweiter Band) has a list of towns on the silk road to the east. The geography of 
Strabo, written about the time of Christ, with several editions in the Loeb Classical Series, as well as the 
Teubner edition by A. Meineke (Leipzig, 1853 foil.), contains much information on Iran and the 
Caucasus. Pliny's Natural History is less valuable, but also has interesting details about lands of the east, and 
both Strabo and Pliny give place names and information older than their own times of writing. 5 

For the Sasanian period the number of geographical sources increases but the quality does not, for the 
various cosmographii in the Geographi Latini Minores give very little information on the east. 6 The 
historian Ammianus Marcellinus, especially in book 23, mentions the provinces and cities of the Sasanian 
Empire. 7 Useful for details about place names and peoples are the lexicon of Hesychius (fifth century) 
where some Persian and 'Scythian' words and names are explained, the geography or Ethnika of Stephan 
of Byzantium (sixth century), and the dictionary of Suidas (tenth century). 8 The anonymous geographer 
of Ravenna (seventh century) based much of his work on the Tabula Peutingeriana of Castorius and does 
not add much on the east. 9 A secondary source on Armenian place names is a useful reference work for 
that region. 10 Talmudic geography is of minimum value to our concerns, although the Talmudic 
dictionary of Aruch contains interesting items, especially for the Sasanian period. 1 ' Syriac sources also are 
not particularly helpful, although the corpus of Nestorian synods with the locations of bishoprics, as well 
as the acts of martyrs, occasionally give geographical information not found elsewhere. 12 

Thus, it is obvious that the historical geographer of pre-Islamic Iran must rely on items of information 
from many and varied sources. Not only the works mentioned above must be considered but the results of 
archaeology as well, which bring new materials for our understanding of the past almost yearly. For 
example, the site of the Parthian capital of Hekatompylos was not established, until a survey of the area 
and excavations showed that it was the Islamic site of Qumis between present-day Damghan and 
Semnan. 1 3 It is obviously impossible to give a bibliography for all the surveys and site excavations which 
have some relevance to historical geography. The yearly survey of excavations published in the journal 
Iran, however, gives a good picture of work in that country since 1965. This is paralleled by publications 
of the annual symposia on archaeological research in Iran organized by the Iranian Centre for 
Archaeological Research in Tehran, since 1 972, as well as the journal Bastan Chenassi va Honar-e Iran, also 
printed in Tehran. 

By far the most detailed sources of geographical information for the ancient as well as medieval periods 
are the geographies in the Arabic or Persian language. Fortunately, a comprehensive survey of the sources 
relevant to western Iran was made by Paul Schwarz, to which one may add a few items which were not 
known to him. 14 The Hudud al-'dlam is perhaps the most important new source, especially for eastern 



5 Plinii Naturalis Historia, ed. by C. MayhofT 
(Dresden, 1905; reprinted Stuttgart, 1967); also 
Loeb Classical Library (Cambridge, Mass- 
London). 

6 A. Riese, ed., Geographi Latini Minores (Heil- 
bronn, 1878; reprinted Hildesheim, 1964) esp. 45, 
61, 74. 

7 Ammianus Marcellinus, ed. by W. Seyfarth, 4 
vols. (Berlin, 1968-71), and in the Loeb Classical 
Series. Both have a translation as well as Latin text. 

K. Latte, ed., Hesychii Alexandrini Lexicon, 3 
vols. (Copenhagen, 1953, 1966, 1977); A. Mein- 
ecke, ed., Stephan von Byzanz Ethnika (Berlin, 
1849; reprinted Graz, 1958); Ada Adler, ed., 
Suidae Lexicon (Leipzig, 1928-35), 4 vols. 

J. Schnetz, Itineraria Romana II: Ravennatis 
anonymi cosmographia (Leipzig, 1940) trans, by 
Schnetz, Ravennas Anonymus: Cosmographia, in 
Nomina Germanica, 10 (Uppsala, 1951). 

H. Hiibschmann, "Die altarmenischen Orts- 
namen," IF, 16 (1904), 197^190; also issued as a 
book in 1969. 

The book by A Neubauer, La geographie du 
Talmud (Paris, 1 868), is still useful, as is A. Berliner, 



Beitrdge zur Geographie und Ethnographie Babylo- 
niens (Berlin, 1884). For the Talmudic geography of 
Mesopotamia the best work is by J. Obermeyer, Die 
Landschaft Babylonien im Zeitalter des Talmuds und 
des Gaonats (Frankfurt, 1929). See also Additamenta 
ad Librum Aruch Completum, ed. by A. Kohut 
(Wien, 1937; reprinted New York, 1955). 

1 2 J -B. Chabot, Synodicon orientate (Paris, 
1 902) ; G Hoffmann, Ausziige aus syrischen Akten 
persischer Martyrer (Leipzig, 1880), and O. Braun, 
Ausgeii'dhlte Akten persischer Martyrer (Kempten- 
Mumch, 1915). 

1 3 J Hansman, "The Problem of Quirm." JRAS 
(1968) 111-40. 

1 •* P. Schwarz, Iran im Mitlelalter (Leipzig, 
1927-36) The book of G. LeStrange, Lands of the 
Eastern Caliphate (Cambridge, 1905; reprinted 
1930), is less detailed but useful. The Istoriko- 
geograficheskii obzor Irana (St Petersburg, 1903, of 
Bartold, reprinted in his Sochineniya, 7, Moscow, 
1971) contains much material of pre-Islamic 
geography, but it is especially valuable for 
Afghanistan, below. The value of Islamic sources 
for ancient history is shown by such publications as 



Geographical Survey 3 

Iran, Afghanistan and Central Asia. 15 The second source, previously known from quotations in Yaqut's 
geographical dictionary and elsewhere, was found in a manuscript in Meshhed, Iran and published 
twice. 16 A later source The Jahan-name contains little new except a few items, such as the building of a 
long wall in Gurgan by the Sasanians to protect settled people from raids by nomads. 1 7 Finally, local 
geographies and histories occasionally give items of interest. The bibliographical references to the most 
important books below refer to pages in Storey's Persian Literature, a bio-bibliographical survey in its 
enlarged Russian translation. 18 The oldest is the history of Qom (p. 1008), which contains information 
about the Arab conquests and divisions of land and irrigation. The Qualities of Isfahan, a similar book, 
existed in Arabic and in an old Persian translation (101 1-1 2), and histories of Yazd (1021), and especially 
the Fats name of Ibn al-Balkhl, which contains much information about the geography of Fars in Sasanian 
times (1027), are to be noted. The city histories of Baihaq and of Nishapur in Khurasan (1041 , 1044), and 
Herat in Afghanistan (1043, 1046) have items of geographical interest, whereas the city histories of 
Kerman are less useful (1056-63). The local history of Tabaristan (present Mazandaran) is important for 
ancient geography and history (e.g. the letter of Tansar), while other books on the Caspian Sea Coast 
provinces are less important (1071-77). The local history of Seistan is an important source book for the 
historical geography, as well as history, of that province (1078-81), but its legends obscure much of the 
history. The same is true of the local history of Shustar in Khuzistan (1082). Azerbaijan and Kurdistan 
have a number of local histories (such as the Sharaf-name, 1097), but they help little with the ancient 
history or historical geography of the respective regions. 

Finally, European travellers from the time of Marco Polo occasionally provide geographical 
information relevant to ancient times not found elsewhere. A survey of such accounts may be found in a 
book by Alfons Gabriel, 1 9 but his book contains errors and has an insufficient bibliography hence must be 
used with caution. There are also general bibliographies of Iran which are useful to consult. 20 
Literature: Afghanistan, general: Parallel to Iran, there exists a gazetteer in Persian as well as a Pashto 
edition, the Qamus Joghrafiya-ye Afghanistan in four volumes and several reprintings in Kabul (first ed. 
1952). The English language parallel is the Historical and Political Gazetteer of Afghanistan, ed. by L. 
Adamec in six volumes, five of which have appeared. 21 A general, physical geography of Afghanistan is 
the work by Johannes Humlum, La geographic de /'Afghanistan (Copenhagen, 1959). On the historical 

M. Streck, Die alte Landschaft Babylonien nach den Vorderasiens," in 10. Internationaler Kongrefi fur 

arabischen Geographen (Leiden, 1900). Namenforschung, (Vienna, 1969), 123-30 

15 Text, ed. by Manuchehr Sotodeh (Tehran, 20 Y. M. Nawabi, ed., A Bibliography of Iran, 2 
1340/1962), trans, with notes by V. Minorsky vols. (Tehran, 1969 and 1971); A. Abolhamid and 
(London, 1937). The MS is now in Leningrad and N. Pakdaman, Bibliographic francaise de Civilisation 
was published in facsimile by V. Bartold (Lenin- Iranienne, 3 vols. (Tehran, 1972-74); A. Kazemi, 
grad, 1930). "Addenda" to the book were pub- Iran Bibliographic Deutschsprachige Abhandlungen, 
lished by V. Minorsky in the BSOAS, 17 (1955), Dissertationen, usw. (Tehran, 1970). In Persian are 
250-270. the valuable Fihrist Kitabhaye chapl Farsl, ed. by 

16 V. Minorsky, ed., Abu-Dulaf Mis'ar ibn Kh. Moshar, 5 vols. (Tehran, 1972-77) an index of 
Muhalhil's Travels in Iran, circa A.D. 950 (Cairo U. printed books, as well as the Fihrist-e maqalat-e 
Press, 1955) and P. G. Bulgakov and A. B. Farst,ed. by I. AfshSr (Tehran, 1961), which is an 
Khalidov, eds., Vtoraya zapiska Abu Dulafa (Mos- index of articles published in Persian from 1910- 
cow, 1960). 58, Two other works in Persian by I. AfshSr are 

17 Fol 1. 19a, line 23 of Muh. b Najib Bakran, useful, the Bibliography of Bibliographies on Iranian 
Dzhahan-name, ed. by Yu. Borshchevskii (Mos- Studies (Tehran, 1964), and Directory for Iranian 
cow, 1960), also ed. by M. Amin RiyShT (Tehran, Studies (Tehran, 1971). Other bibliographies exist 
1964), 82. m profusion, but noteworthy is W. G. Oxtoby, 

Storey's work was published in London in Ancient Iran and Zoroastrianism in Festschriften 

fascicules (1927-72), and the enlarged translation (Pahlavi University, Shiraz, 1974), and the Russian 

Persidskaya Literatura, ed by Yu. Bregel and Yu. language Bibliografiya po geografii Irana by M. P. 

Borshchevskii in 3 vols. (Moscow, 1972). Petrov (Ashkabad, 1955), an annotated 

A. Gabriel, Die Erforschung Persiens (Wien, bibliography. 

1952). There is not space here to mention many 2I The three provinces or areas covered in the 

special, detailed studies on historical geography, three published volumes are Badakhshan (Graz, 

but the articles by W. Eilers on this subject should 1972) Farah and Southwestern Afghanistan (Graz, 

be noted, "Der Name Demawend," /lO, 22 (1954), 1973) and Herat and Northwestern Afghanistan 

267-374 and 24 (1956), 183-224, and "Zur (Graz, 1975), all compiled from British-Indian 

Ortsnamengebung und Ortsnamen-Forschung gazeteers. 



4 Chapter I 

geography of the area, the same sources mentioned above for Iran also apply to the east, but the works of 
Markwart, notably his Wehrot und Arang (Leiden, 1 938), are particularly useful for this part of the world. 
A special study of Ptolemy's geography by Italo Ronca, and the relevant chapters of Bartold's Istoriko- 
geograficheskii obzot Irana are noteworthy. 22 Also V. Bartold's Turkestan down to the Mongol Invasion 
(London, 1968) covers northern Afghanistan as well as Central Asia. 

Sources: In addition to the Classical sources listed for Iran, especially Herodotus, Strabo, Pliny and Isidore, 
particular mention must be made of the historians of Alexander's campaign, for they give a few items of 
geographical information not found elsewhere. 23 Unfortunately, Indian texts give us no help, and even 
the geography of Varahamihira, except for names of tribes in the northwest of the subcontinent, is too 
vague to be of much value. 2 * More informative are Chinese sources, especially the life and the travels of 
Buddhist monks or laymen such as Fa-hsien (c. 400 B.C.), Sung-yiin (c. 518) and especially HsOan-tsang 
(629 B.C.). 25 These travel accounts, however, frequently give us names which are difficult to identify in 
non-Chinese texts. In addition to the accounts of travellers, the dynastic histories of China also have a 
modicum of geographical information about lands comprising present-day Afghanistan. Most of the 
sections about Iran, Afghanistan and Central Asia in those texts fortunately have been translated into 
European languages thus making them available to those who do not read classical Chinese. The oldest 
account comes from an embassy sent by the Chinese court starting about 138 B.C. headed by Ch'ang 
Ch'ien, whose account was incorporated in the history of the former Han dynasty (Ch'ien Han Shu). 26 
The annals of the later Han dynasty, the Hou Han Shu, do not add much to the account of the western 
regions in the annals of the former Han. 27 Later dynastic histories such as the Wei-Ho and the Chou Shu, 
give additional information. 28 Even later Chinese sources are still useful for the earlier historical 
geography of the western regions (in Chinese eyes), such as the voyage of another Buddhist monk Huei- 
cha o about 726 a.d., and the later dynastic histories of the Sui and Tang. 29 

The same geographical books in Arabic and Persian, mentioned above, also contain information about 
Afghanistan, but the Hudttd al-a'lam is especially detailed. Since the mountainous regions of Afghanistan 
were converted to Islam long after the Arab conquests in Iran, several later histories relating to 
Afghanistan are more valuable for the earlier historical geography than one would expect. These are the 



22 Italo Ronco, Ptolemaios, Ceographie 6, 9-21, 
Ostiran und Zentralasien (Rome, 1971). See also A. 
Berthelot, L'Asie ancienne centrale el sud-orientale 
d'aprh Ptolemke (Paris, 1930). 

23 The principal sources are Arrian, Diodorus 
Siculus and Quintus Curtius, all available in either 
the Loeb or Teubner series of Classical texts. In the 
appendices to W. Tarn's The Greeks in Bactria and 
India (Cambridge, 1951), many geographical 
questions are discussed. 

24 H. Kern, ed., The Brihat Sanhita, Bibliotheca 
Indica, 48 (Calcutta, 1865), also ed. by S. Dvivedi 
(Benares, 1895-7), Sanskrit Brihatsamhita . Trans, 
into English by Kern in JRAS, new series 4 (1 869), 
430-79; 5 (1870), 45-90, esp. ch. 14 on the 
division of the globe 81-86; 6 (1872), 279-338; 7 
(1874), 81-134. 

25 Classical Chinese texts are available in many 
printings and only translations are mentioned here. 
For Fa-hsien and Sung-yun see S. Beal, Travels of 
Fah-hian and Sung-yun (London, 1869), also his 
translation of 77ie Life of Hiuen-Tsiang (London, 
191 1). The best translation of Sung-yun, however, 
is by E. Chavannes in the BEFEO, 3, no. 3 (Hanoi, 
1903). For Hstlan-tsang's travels the translation by 
T. Watters, On Yuang Chwang's Travels, 2 vols. 
(London, 1904) is better than Beal's (London, 
1884). 



26 The Ch'ien Han Shu, relevant chapters have 
been translated by J. J. deGroot in Chinesische 
Urkunden zur Geschichte Asiens, die Westlande 
Chinas in der vorchristlichen Zeit (Berlin, 1926), 9- 
19. Also trans, by F. Hirth, "The Story of Chang 
K'ien," JAOS, 37 (1917), 89-162. 

27 E. Chavannes, "Les pays d'Occident d'apres le 
Heou han Chou," TP (1907), 149-234. The 
translations of Iakinf Bichurin into Russian of this 
and other Chinese accounts are outdated: N. Ya. 
Bichurin, Sobranie svedenii o narodakh obitavshikh v 
Srednei Azii, 3 vols. (Moscow, 1950-53) also his 
Sobranie svedenii po istoricheskoi geografii vostochnoi 
i srednoi Azii (Cheboksarai, 1960), both reprints. 
The same is true of the old French translations of 
Chinese texts by S. Julien. 

28 E. Chavannes, "Les pays d'occident d'apres le 
Wei-lio," TP (1905), 519-71; R. A. Miller, 
Accounts of Western Nations in the History of the 
Northern Chou Dynasty (Berkeley, 1959). 

29 For the monk see W. Fuchs, "Huei-cha o's 
Pilgerreise durch Nordwestindien and Zenral- 
Asien um 726," Sb PAW, 30 (1938), 426-69. For 
the histories, the collection and translations in E. 
Chavannes, Documents sur les Tou-kiue (Turcs) 
Occidentaux (St. Petersburg, 1900; reprinted Paris, 
1946), esp. the discussion of sources on the 
"western countries", pp. 99-100, is very useful. 



Geographical Survey 5 

history of Gardlzland that of JazjanT, both of whom were natives of the area. 30 There are no extant local 
histories of Kabul, Qandahar, Ghazna, Bamiyan or other cities, although Balkh has a local history which 
gives very little ancient history or topography of the surroundings. 31 Finally, European travel accounts 
occasionally contain items of value for the historical geography of the land. Archaeology has been the 
monopoly of the French until after World War II, and the series MDAFA contains much valuable 
information, while the latest information about archaeology may be found in the journal Afghanistan, 
published in English or French by the Historical Society of Afghanistan. 

Several bibliographies of Afghanistan exist which are useful for specific references in modern literature, 
and together they offer good coverage of writings on the land. 32 

Literature: Central Asia, general : There are no gazetteers as above, but rather lists of place names as in B. 
Volostnov, Slovar geograficheskikh nazvanii SSSR (Moscow, 1968) and S. V. Kalesnik, Entsiklopedicheskii 
Slovar geograficheskikh nazvanii (Moscow, 1973). For each of the Transcaucasian and Central Asian 
republics of the USSR there exists a general geography, in the series Souetskii Soyuz, and some of these 
books have been translated into English in abbreviated form, but only the Russian originals should be 
consulted; for Uzbekistan, see Korzhenevskii, N. L., Uzbekskaya SSR (Moscow, Geografgiz, 1956); for 
Turkmenistan, Z. G. Freikin, Turkmenskaya SSR (Moscow, 1957). and for Tajikistan, I. K. Narzikulov, 
Tadzhikskaya SSR (Moscow, 1956). 33 For older works see the bibliography in I. M. Kaufman, 
Geograficheskie Slovari i bibliografiya (Moscow, 1964). 

Sources: The same Classical sources which have been mentioned above are also relevant for Central Asia, 
especially Herodotus, Strabo and Ptolemy. Particularly to be noted for this area is Ctesias, who constantly 
must be checked, however, with other sources. 34 Among secondary publications, the works of Markwart 
are indispensable for anyone studying the historical geography of Central Asia. In addition to Wehrot und 
Arang (Leiden, 1938), the continuation of that book, printed in two articles, as well as an earlier study of 
Tomaschek, are noteworthy. 35 

The Islamic and Chinese sources for the historical geography of Central Asia are the same as those 
mentioned for Afghanistan. The pioneering work of V. Bartold, Turkestan down to the Mongol Invasion, as 
well as Minorsky's notes to his translation of the Hudud al-'alam, are both very informative. For 
archaeology in Central Asia and the Transcaucasus, the yearly reports of field work, Arkheologicheskie 
Otkrytiya, ed. by B. A. Rybakov for the year 1975 (Moscow, 1976), as well as the journal Sovetskaya 
Arkheologiya, are indispensable reference works. Most of the republics have their own surveys of 
archaeological work done within the borders of the republic every year, as, for example, Arkheologicheskie 
Raboty v Tadzhikistane, vypusk II (Dushanbe, 1975). Other publications of value for the geography of 
Central Asia are the series of the Vostochnaya Komissiya geograficheskogo Obshchestva SSSR, called Strany i 
Narody Vostoka, vol. 1 6 of which (Moscow, 1 975) is devoted entirely to the Pamirs, also the Trudy and the 
Materialy of the Khorezmskoi Ekspeditsii and the same for Turkmenistan in the Yuzhno-Turkmenistanskoi 

30 'Abd al-Hayy Hablbl, ed., Zain al-akhbar of 33 There are other books in series with useful 
GardTzT (Tehran, 1969), esp. the final chapters; he information, such as the general survey of each 
also edited the Tabaqat-e Naslrl of JnzjSnT (Kabul, republic, e.g., D. A. Chumichev, Tadzhikistan 
1965) in two vols., the second with notes. (Moscow, 1968), on climate, physical features, 

31 See the Fadai'l-e Balkh, ed. by 'A. HablbT production, etc. Also the Atlas Taizhik SSR 
(Tehran, 1972). The contemporary book on (Moscow, 1 968) has historical, as well as contem- 
Qandahar by Muh. Wali Dzalmai (Kabul, 1973) porary, maps. The Atlas of Uzbekistan, on the 
does give geographical information, but it is other hand, is quite poor. 

written in Pashto. 34 See F. W. KOnig, Die Persika des Ktesias von 

32 T. I. Kukhtina.ed., Bibliografiya Afghanistan, Knidos (Graz, 1972), and several articles by I. V. 
literatura na Russkom Yazyke (Moscow, 1 965) ; W. Pyankov in the VD1, such as "Svedeniya Ktesiya o 
Kraus, ed., Bibliographic der Afghanistan-Literatur vladeniyakh Bardii na vostoke Irana," no. 4 (1961), 
1945-1967, Teil 1, Literatur in Europaischen 98-103, "Istoriya Persii Ktesiya i sredneaziatskie 
Sprachen (Hamburg, 1 968), and Teil 2, Literatur satrapii Akhemenidov," VDI, no. 2 (1 965), 35-50. 
in persischer Sprache und Paschtu, ed. by G. Knabe 35 Markwart, "Die Sogdiana des Ptolemaios," 
(Hamburg, 1971); D. Wilber, Annotated Bibli- Orientalia, 14 (1945). 123-49,and 15 (1946),286- 
ography of Afghanistan (New Haven, 1968); E. 323; W. Tomaschek, "Zentralasiatische Studien," 
Naby, Recent Books about Afghanistan 1968-1973 Sb WAW, 87 (1877). 

(Afghanistan Council, The Asia Society, New 
York, 1973). 



6 Chapter I 

arkheohgicheskoi kompleksnoi ekspeditsii. 36 Central Asia is also well-represented in the series Matetialy i 
issledovaniya po arkheologii SSSR. The vast field of archaeology, of course, cannot be covered and 
individual publications will be mentioned in the succeeding chapters relative to particular questions. 

City histories are also of value for the historical geography of these lands, and the most famous in 
Central Asia is the history of Bukhara by Narshakhi in Persian (Storey-Bregel, 1108), while the 
counterpart for Samarqand, known as al-Qandiyya (1112) does not contain pre-Islamic information as 
does Narshakhi. Other local histories of Bukhara or Samarqand are too recent to have information 
relevant to ancient history or geography. The memoirs of the first Moghul emperor, Babur, contain a 
geographical sketch of the Ferghana valley and surroundings (828, 1 187). Other texts, unfortunately, are 
of little or no value for ancient times. 

A word should be added about maps, since now very detailed aerial maps exist for most of Iran, 
Afghanistan and Central Asia, but, unfortunately, access to them is frequently difficult. For example, the 
photomaps of Afghanistan, prepared for the government of that country by Fairchild Aerial Surveys of 
Los Angeles (or Teledyne Geotronics), have a scale of 1 : 50,000 and show surprising details, but they are 
not available for general use. A stereo-topographic survey of Afghanistan in 1959, produced maps of 
1 :100,000, printed in 1960. For ordinary purposes, however, maps of 1 :1 ,000,000 are adequate, and these 
exist in many editions for the entire area. The various Soviet republics have such maps, published by the 
Glavnoe Upravlenie geodezii i kartografii gos. geologicheskago komiteta SSSR, in Moscow in the 1960's, 
while the U.S. Army Map Service, Corps of Engineers has available a series of world road maps. The 
German Army maps, made in the 1930s are also very good for topographical details over the entire area. 
Unfortunately, no good atlases or series of maps on the ancient history of Iran, Afghanistan or Central 
Asia exist, and many ancient names cannot be identified with present sites, but at least we are much better 
informed about geography today than only a few decades ago. 



THE ENTIRE AREA 

It is no exaggeration to say that rugged mountains and barren deserts have 
conditioned the lives of Iranians on the Iranian plateau and in Central Asia perhaps 
more than have similar conditions for other peoples. For the mountains and deserts 
have divided and separated the people more than rivers or lakes, and consequently 
many peoples, many languages and dialects and ways of life have developed in isolated 
valleys. Geography is vitally important for a study and understanding of this part of 
the world. A common feature of most of the areas under discussion here (modern Iran, 
Afghanistan, Turkmenistan, Uzbekistan and Tajikistan) is that they all are on a 
plateau or series of plateaus, the middle parts of which have the lowest altitudes, while 
the highest mountain ranges are on the peripheries. Rivers and streams drain into the 
central depressions, frequently forming salt lakes or dried lake beds which are deserts 
today. The greatest contrast in height and depth is found between the highest peak of 
the Elburz range, Mt. Demavend (5604 m.) and the Caspian Sea, less than 100 km. to 
the north, which is below sea level. Both the Elburz range in the north and the Zagros 
range to the west, branch off from the Caucasus in the northwest of Iran, the former 
extending to the east and the latter to the southeast, with many valleys and mountain 
offshoots. In the east the Elburz range becomes lower with one series of mountains 
continuing into Afghanistan, while another branch, of less altitude, extends south to 
the depression of Seistan. 

The main characteristic of the Iranian plateau is the drainage of rivers into salt lakes 



36 The southern Turkmen expeditions have a tematike yuzhna-Turkmanistanskoi arkh. komp. eksp. 
bibliography of writings about them from 1949- Akademii Nauk Turkmen. SSR (Ashkhabad, 1970). 
69, Perechcn ' opublikovannykh rabot i materialov po 



Geographical Survey 7 

or into the sands in the interior. Other than the Caspian Sea, the largest lake in the 
world, the salt lake of Urmia (formerly Rezaiye) is the main place of drainage for 
streams in Azerbaijan. Other lakes are seasonal, spreading over large areas at the end of 
winter and almost vanishing at the end of summer. Such are the Darya-ye Namak 
(Salt Lake) east of Qom, the fiakhtegan lake south of Persepolis and Maharlu lake 
south of Shiraz. In the east, the Hamun in Seistan receives the water of the Helmand 
River, while other seasonal lakes are much smaller. It is possible that in antiquity all 
over Iran more water was available than in recent times, and certainly in many 
regions, such as the mountains of Fars, the vegetation and number of trees were 
greater than today. But on the whole the land is much the same today as it was 3000 
years ago, in both temperatures and landscape. Water and vegetation are and were 
more plentiful in Azerbaijan than elsewhere, except on the Caspian Sea coast, and the 
land becomes more arid as one goes south and east, where extensive irrigation was and 
is necessary for cultivation. The creation and extension of the underground canal 
system for irrigation, called qanat in Iran and karez in Afghanistan and Central Asia, 
developed probably at the beginning of the first millennium B.C., according to some 
archaeologists, enabled people to move out upon the plains from mountain valleys, 
thus enabling a larger population to develop. 37 It is conceivable that the spread of the 
Iranians on the plateau was aided by the qanat irrigation system, but in any case the 
spread of both occurred roughly at the same time. 

Afghanistan and Central Asia have much the same geographical features as Iran, 
except the eastern part of Afghanistan, which lies in the drainage zone of the Indus 
River basin. Yet even here, in the relation of the mountain highlands of the 
Hindukush range to the plains of India, we have a parallel with the Zagros range in 
the west and the plains of Mesopotamia. Internal lakes, into which streams drain, are 
fewer, smaller and generally higher in altitude in Afghanistan than in Iran. In that 
part of Central Asia where Iranians in the past established extensive settlements 
(present Turkmenistan, Uzbekistan and Tajikistan), rivers, on the whole, drained into 
the Caspian or Aral Seas. North of these three areas streams fed the seasonal Lake 
Balkash and the deep Issyk Kul. The vast steppe lands to the north, however, placed 
limits on Iranian settlement in that distant region. The general picture of the entire 
area of Iran, Afghanistan and Central Asia is the same for all, mountains on the edges 
and in the middle lands sloping from the mountain ranges, with drainage into 
internal seas, lakes or marshes. Deserts and general aridity also characterize the entire 
area, which, for the most part, has a continental climate, except for the narrow bands 
of land which are adjacent to the Caspian Sea, the Persian Gulf, or the plains of India. 
Since geography has played such an important role in the history of this part of the 
world, a more detailed local survey is appropriate. We start with the area of present 
Iran, then Afghanistan, and finally Central Asia. 



7 On the origins of the important qanat system, Annates: honomies-societe's-cwilisations, 18 (Paris, 

see P. W. English, The Origin and Spread of 1962), 499-520. B. Brentjes, "Studien zum Bewas- 

Qanats in the Old World," Proceedings of the serungsackerbau," Schriften zur Geschichte und 

American Philosophical Society, 112 (Philadelphia, Kultur des alien Orients, 11, AF, 1 (1974), 53-4, 

1 968), 1 70-81 , and H. Goblot, "Dans l'ancien Iran, suggests that qanats are an Urartian discovery, 
les techniques de l'eau et la grande histoire," 



g Chapter I 

IRAN 

Since Persis (OP Parsa, NP Fars) was the center from which both the Achaemenid and 
the Sasanian dynasties arose, we can start with the geography of this part of Iran. Fars 
can be described best as a series of steps leading from the Persian Gulf to the central 
deserts of Iran, with the largest and most fertile step or plain being that of present 
Marvdasht where the ruins of Persepolis stand. It was in the rich valleys running 
northwest to southeast between ridges of the Zagros range, that the human resources 
were found several times to expand and create empires. The average annual rainfall of 
350 mm. at Shiraz, compared with over a thousand mm. along the Caspian Sea coast, 
indicates the relative aridity of the south. Nonetheless, the altitude of the mountain 
valleys, plus water from melting snows provided sufficient water for irrigation, thus 
enabling a sizable population to live in the broad valleys. The mountains between 
Pasargadae and Abadeh (the latter still included in Fars) in the past have provided a 
barrier for communications with the north. The salt and swamp depression, leading 
from east of modern Abarqu southeast to the pass east of Darab, also provided a 
natural barrier isolating Fars from the rest of the plateau. The boundaries of Fars, of 
course, varied at different times because of political events, more than sketched above, 
but the geographical boundaries can be determined as encompassing the land between 
the Persian Gulf and the mountains on the north and the salt depression in the east. 
The salt depression, of course, was nothing like the central deserts of Kavir and Lut, 
but it was nonetheless an inhospitable band of land varying from 50 to 100 km. in 
width and descending to 400 m. above sea level east of Darab. 

Settlements in Fars, as elsewhere on the plateau, were primarily in oases where 
water was obtainable for raising vegetables and gardens. These settlements were 
either in the warm district (garmslr) or in the cold district (sardslr), according to the 
classical Islamic geographers. By this they meant that a settlement could grow either 
tropical plants or trees (such as date palms) or temperate ones (apples, pear trees). The 
oases, of course, could be large or small, but the character of an oasis is determined by 
the availablility of water. In some plains or broad valleys a succession of small 
settlements or oases flourished, while in others water was led to a central town, and 
throughout history the site of this town might vary from place to place. 

The Marvdasht region is really a series of connecting plains separated by individual 
mountains but not by ranges. In other words, one could travel from Malyan tepe (in 
the Baida district), the site of the ancient Elamite city of Anshan, to Persepolis, 
continuing to the Bakhtegan salt lake in the south without crossing any mountain 
pass. To go to Pasargadae from this plain, however, meant following a river, at times 
through gorges, to climb to a higher altitude where it was much colder in winter. 
The plain and rolling hills around Pasargadae were better suited for shepherds than 
for settled folk, and this region, in any case, was the last cultivated area before one 
reached the mountains of Fars to the north, which were not high but wide and with 
little vegetation. From Marvdasht (alt. 1737 m.) to the west one descended, passing 
over mountain ranges, to the next and lower step on the plateau, the present Shiraz- 
Kawar-Khafr-Jahrum series of plains going south, or Shiraz-Guyum-Ardakan to the 
north. Again no high mountain passes break up this chain of valleys or plains, and the 
drainage of water is also into a salt lake, called today Maharlu, to the south of Shiraz. 



Geographical Survey 9 

The altitudes of settlements also are lower towards the south, from c. 2212 m. at 
Ardakan to c. 1600 m. at Shiraz and 1065 m. at Jahrum. Three side plains were 
located southeast of Maharlu lake, that of Sarvistan (alt. c. 1645 m.), Fasa (c. 1560 m.) 
and Darab (c. 1 1 88 m.). The altitudes are important to show the descent from a higher 
northern part of the step to a lower one. The next step towards the Gulf was down to 
the plains of Firuzabad (c. 1350 m.), Kazerun (c. 815 m.), Fahliyan (886 m.) and even 
lower, Behbehan (335 m.), actually in Khuzistan province, where the greatest 
altitude, unlike the other steps, is reversed, descending from south to north. After this, 
we find mountain ranges going down to sea level on the Persian Gulf. Only the 
coastal plain of Borazjan-Bushire is wide enough for extensive settlements, 
depending primarily on the date palm as their major agricultural product. In the 
south of Fars a few isolated valleys, which have never supported a large population, 
have served as way stations for merchants travelling from the Gulf to northern 
centers. The most important valley is that of Khunj-Ivaz— Lar (909 m.) where even 
today local dialects are spoken, an indication of its isolation from the rest of Fars. Thus, 
Fars itself is isolated on all sides, from the mountains rising from the Gulf to the great 
altitude drop in the Bandar 'Abbas depression east of Darab and the salt desert band 
east of Abarqu and north and east of Darab. The cohesion of this part of Iran, seen also 
in the dialects which constitute a unity with modern Persian, strikes the traveller in 
comparison with other parts of the country. None of the rivers here are navigable, 
and mountains dominate the landscape. Also the people, especially in villages, seem 
less variegated than elsewhere in Iran, probably the result of the isolation. Finally, 
because of the variety of climate, altitudes and large tracts fit only for herding, 
traditionally Fars has been a province with the greatest variety of migrating nomads 
in all of Iran. Nomadism seems to have flourished here from early times, and the 
relatively short distances from summer to winter quarters, just as on the eastern rim of 
the plateau in Afghanistan and the plains of India, have encouraged the development 
of nomadic life. 

The same situation, of course, also obtained in the present day province of 
Khuzistan where nomads moved in the summer from the plains of Mesopotamia 
onto the Zagros range, and this has been a constant feature of the history of this 
province. The navigable Karun River, widest in Iran, flows through the province, but 
it must be remembered that the land to the south of Ahwaz, to the Gulf, was not 
occupied in historic times, except along the river banks, whereas north of Ahwaz the 
land was fertile and heavily cultivated. A number of tributaries of the Karun make 
Khuzistan a well-watered province. Here was located the Elamite city of Susa, and 
other ancient towns, and since this region is an extension of the plains of Mesopotamia 
to the east, influences from the west always have been strong. 

The wide mountainous tract to the north and east of Khuzistan is Luristan, where 
the population is and was different from the people on the plains. Luristan, however, 
was not like Fars, with wide and long plains between the mountains, but rather many 
small and narrow valleys with a variety of settlements separated from each other. 
Also roads pass through Luristan, connecting the plains with the plateau, corridors of 
trade. Thus, the geography of Luristan, unlike Fars, was not conducive to the unity of 
the people inhabiting this large area. Furthermore, control of Luristan from the 
outside was always difficult and nominal. 



JO Chapter I 

The only centers of population which could expand, were in valleys on the edges of 
Luristan, such as present Borujird. Khurramabad, on the other hand, was important 
rather for its strategic location on a trade route to the plains of Khuzistan. Settlements 
might vary in location from age to age, but there was not much livable space between 
the mountains for great displacements of people. Even as today, Luristan in the past 
provided summer quarters for the inhabitants of the plain seeking to escape the 
excessive heat. 

We find a similar situation in the northern Zagros mountains, in Kurdistan and 
Azerbaijan, while the ancient road from Babylon to Khurasan in the east, sometimes 
called the 'Silk Route,' was a dividing line between Luristan and Kurdistan. The latter, 
comprising the mountainous areas from Kermanshah north to Lake Urmia at present 
shows an interesting linguistic division between Kurdish and Turkish speakers. In the 
mountains to the west of the valley of the Simin and Zirineh Rivers, where the 
present-day towns of Mahabad, Saqqiz and Bijar are found, the population is Kurdish- 
speaking, while in the lower areas, the line of towns from present Miyandoab south 
through Takab, and in the mountains to the east, the population is Turkish-speaking. 
It is doubtful whether this present split represents an ancient division rather than 
being the result of political and other factors. One may rather consider the entire area 
of Kurdistan and Azerbaijan as a geographic whole, with more of a significant 
geographical division existing in the mountains between the Van and Urmia lakes 
than between the valleys inside of Azerbaijan. Geographically speaking, lake Urmia 
can be considered a link rather than a barrier between the eastern and western parts of 
Azerbaijan, and the land along the western shore of the lake from Urmia through 
Salmas, then north to Khoi and Maku presents no great geographical barrier to the 
movement of peoples. It is curious that, regardless of political controls and divisions, 
the ethnic boundary between Armenians and Iranians, in Sasanian times if not earlier, 
was the line of hills separating the present district of Urmia from Salmas, and 
generally this seems to have been the border between the Urartians and the Manneans 
before the coming of the Armenians and Iranians. 

It must be remembered that routes of communication in antiquity did not 
necessarily follow modern highways, since wheeled vehicles were very little in use in 
Iran; instead pack animals, such as donkeys, mules and camels, carried merchandise 
over mountain passes and plains. Nonetheless, the pack animals followed the lines of 
least resistance, and the modern road from Mt. Ararat and Maku in the northwest 
through Tabriz, Mianeh, Qazvin, Tehran, Semnan and to Khurasan, was the main 
trade artery together with the branch which descended from the plateau at Hamadan 
through Kermanshah to the plains of Mesopotamia. Since these were also the routes of 
invasion and conquest from earliest times, the populations along these routes were 
variegated and did not present the same unity of peoples along them as, for example, 
the inhabitants of Fars. Control of the 'Silk Route' to the east and of the branch 
through Azerbaijan to the Black Sea, would ensure control of most of the Iranian 
plateau, and this control was necessary to maintain an empire, or even unity over the 
plateau. In Azerbaijan present settlements along the route, such as Marand, Tabriz, 
and Mianeh, were also ancient centers, showing the continuity of routes from 
antiquity. Other towns such as Maragheh, Ardabil and Ahar are not on important 
routes of communication but owed their importance on the whole to agriculture. It is 



Geographical Survey li 

richer in Azerbaijan than elsewhere on the plateau because of the relative abundance 
of underground water, as well as streams from melting snow on the mountains, not to 
mention the soil itself, in many places lava from now extinct volcanoes. 

One area of Azerbaijan is different from the rest, and that is the low-lying steppe of 
Mughan, which served in later times as a winter camping ground for the Mongols 
and Turkish nomads. This lowland, which continues north across the Aras River, was 
the division in the northeast between Iranian speakers and the land and people known 
as Aghvan to Armenians, Arran to the Arabs and Albania to the Greeks. Since this 
was non-Iranian land in the north we need not describe it. The eastern boundary of 
Azerbaijan was not the Caspian Sea but rather the high mountains along the coast 
with a narrow band of coastline which is part of the Caspian provinces, with a 
landscape very different from the rest of Iran. Those provinces are Talysh, Gilan and 
Mazandaran. 

The coastal area of Gilan, including Iranian Talysh, is roughly 225 km. long from 
northwest to southeast and the width varies from 1 5 km. or less in the Talysh area of 
Astara to over 100 km. by Rasht. The actual plain, however, excluding foothills, is 
even narrower, varying from a few hundred meters near Astara to 45 km. at Rasht. 
On the Caspian side of the Elburz mountains are many trees, thick underbrush, called 
locally a jangal. To the east both water and vegetation decrease, but many streams 
descend from the mountains to the Caspian, making communications along the shore 
difficult. Just as in the west the delta of the Safid Rud and its tributaries provided a flat 
agricultural area for development, with its center at Rasht, so in the east the Haraz and 
Babol Rivers with many tributaries made the plain of Mazandaran a center of 
population. But in the east, unlike the west, the plain continues further, to the east of 
the Caspian Sea, forming the wider plain of Gurgan, ancient Hyrcania. Mazandaran 
and Gurgan were linked in history, and both were open to invasion from the steppes 
of Central Asia. The jungles of well-watered Gilan and Mazandaran give way, 
however, to the forests of the hills of Gurgan, and the mountains to the south did 
shield the inhabitants of the Caspian provinces from penetration from the south. 
Archaeology has uncovered remains of ancient cultures in sites such as Marlik, 
Kaluraz and Kalardasht, located in the northern foothills, or even in the higher valleys 
of the Elburz mountains. It is unlikely that the marshes and jungles of the plains were 
much inhabited in early times, but the hills were fertile and watered, enabling people 
to live there. It has not been possible to determine by archaeology when the Caspian 
plains were settled, but probably by the Achaemenid period settlements were being 
established on the plains. Much like India, the lowlands of the Caspian were invaded, 
but the reverse, an expansion from the lowlands to the mountains and onto the 
plateau, was rare. 

The southern sides of the Elburz mountains on the plateau are the opposite of the 
north where winds drop their moisture, as they try to cross the lofty range. The 
aridity of the southern slopes is progressive, however, for the further south one goes 
from the Elburz the less moisture is found. Therefore, the band of land from the 
mountains, one might say from Hamadan through Qazvin, Tehran to Semnan and 
Damghan, is, and has been, the most practical route of communication from west to 
east. Settlements are found on the route where water is available, and this is the main 
reason for the continuous existence from ancient times of a town at Hamadan, where 



12 Chapter I 

springs and streams from Mt. Alvand bring water. Likewise at Qazvin not only do 
streams bring water from the nearby mountains, but the strategic location as a 
crossroads to the Caspian provinces, to Azerbaijan, to Hamadan and to the east, 
insures the continuous existence of a town from antiquity. Further, at Semnan but 
especially at Damghan, which is probably a much older site, the presence of water 
enables settlements to flourish. North of Damghan the copious spring, today called 
Chashmeh-ye 'Ali, undoubtedly provided water for early settlement on the plain, 
first at Hekatompylos (present ruins of Qumis) and later at Damghan. Thus along the 
trade route to the east, access to water provided the means of settlement. 

The same is true of the road south to Fars along the edge of the Kavir salt desert. At 
Kashan the spring of water at Fin, in the mountains to the southwest of the present 
city, provided a similar basis for ancient settlement as at the area of Damghan. On the 
edge of mountains with streams leading down to the central salt depressions of Iran, 
ancient man established himself. The large oasis of Isfahan, however, was watered by 
the Zayandeh River. All of the principal settlements were and are about 1000 m. in 
altitude or higher, for the depressions where streams on the interior of the plateau 
debouched are salt marshes where life is not possible. The main settled part of the 
western Iranian plateau, comprising the area of Hamadan-Qazvin-Tehran-Isfahan, 
was frequently under a single political administration, since there were no great 
obstacles to communication between these cities. The southeastern part of Iran, 
however, is more difficult of access, and is also more arid. 

The same pattern of settlement prevailed here as in the north; wherever streams 
descended from the mountains, or springs existed, settled human life was possible. 
The mountains of the center of the land, however, are not as high as the Elburz or 
Zagros ranges. The present city of Kerman lies 1860 m. above sea level and is 
separated from the Lut desert to the north by a range of mountains, while a former 
capital of the province, Sirjan, is 1650 m. high. Because of the altitude, summers are 
not so intolerable as one might expect in an area in the south. Because it is south, 
however, in the lowlands date palms, mango trees and other tropical plants may be 
found. East of present day Kerman city, however, the land slopes to 1000 m. at Bam 
and then much lower to the east of Bam in the desert sands. Historically, as well as 
geographically, the Lut desert and the sands to the east of Bam have divided the 
Iranian plateau into its western and eastern parts, the latter of which continues into 
present Afghanistan and Pakistan. The only parts of the eastern part of the modern 
country of Iran which sustained settled populations in the past are Seistan and the 
Bampur valley of Baluchistan in the south. Seistan is the large plain into which the 
Helmand River from the mountains of Afghanistan empties, and it lies under 1000 m. 
above sea level, and in parts even down to 300 m. The wind is especially strong here 
and a local summer wind of '120 days' shapes the sand and creates dunes. But the 
overflow of water from the Helmand in the spring creates large reed marshes or 
shallow lakes, called hamun or "plain." The number and size of these hamuns varies 
according to the season and also depending on the amount of water brought by the 
Farah and Helmand Rivers from the snow on the mountains. The Seistan depression is 
large in area, most of which lies within the borders of present Afghanistan. To the 
north of the Seistan depression, and east of Qaen, is a similar but considerably smaller 
depression called the Namaksar where salt marshes or lakes block communications 



Geographical Survey J 3 

between east and west. Here, as usual on the plateau, settlements are at high altitudes to 
the west of the depression, with Birjand 1490 m. and Turbat-e Haidariyeh 1370 m. It 
is clear that human habitation in antiquity as today flourished primarily on streams 
on the slopes of mountain ranges, for the drainage basins of small streams were saline. 
The large basin of Seistan, however, was an exception, since the water from rivers was 
ample enough to provide a source of livelihood in fish and fowl, as well as water for 
irrigation. Nonetheless, Seistan always must have fluctuated between years of plenty 
and years of scarcity, with salinity an ever present threat to overcome. 

A similar situation may be found in Baluchistan where the drainage of several 
small rivers, such as the Bampur River, is into the salt lake of Jaz Muryan. In 
antiquity, as at present, settlements were found along the rivers, the Halil River to the 
west, descending from the Kerman mountains, and the Bampur in the east. In neither, 
however, did any centers of population exist, since the land was too barren and 
unproductive to sustain many people. The Makran mountains down to the coast are 
about as unhospitable a landscape as may be found anywhere on earth, and not until 
the Indus River Valley in the east can any extensive land for cultivation be found. One 
may characterize all of Baluchistan, from the Kerman highlands to the Indus River, as 
a kind of refuge area where people retreated, either driven out of more favorable land, 
or simply fleeing to maintain a free and unattached life, albeit one of great hardships. 

The only trees to be found are palm trees in oases, together with mangroves, and 
occasionally tamarisk and other bushes elsewhere. The climate is exceedingly hot in 
the summer and is not conducive to any human activity. The coastal plains, varying 
in width from a few kilometers up to 100 at Bushire and a similar extent at Bandar 
'Abbas and Minab, usually have had a minimum of contact with the interior, but 
rather communication has been across the sea. Fishing and date palms provide 
sustenance for the small population able to eke out a livelihood from the 
surroundings. The mountains and deserts which separate the coasts from the towns on 
the interior of the plateau are formidable barriers to commerce and communications, 
and only at Bushire and at Bandar 'Abbas have seaports existed which served the 
hinterland as outlets for trading goods. 

Just as the southern ports have served as entrepots for seaborn trade, so on the 
northeastern frontier of Khurasan towns on the edge of the desert have served as 
places of entry for nomads and merchants with caravans coming across the Kara Kum 
desert. The valleys between the two ranges of mountains, the Turkmen or Kopet 
Dagh range in the north, forming the present boundary between Iran and 
Turkmenistan, and the southern Khurasan or Binalud range north of Nishapur, are 
relatively fertile and well watered by underground canals from the mountains. 
Human settlement in this region is old, and locations of ancient towns are almost 
predictable, as for example, at Tus or Meshhed, or at Isfarain, a fertile valley in the 
mountains. The mountains of Khurasan are not high and passes through them are 
frequent. On the northern slopes of the Turkmen range lies the ancient Parthian city 
of Nisa, near present day Ashkabad, and seemingly from ancient times a settlement in 
this region (as at Anau) has served as a center of trade along an east-west axis. 
Southern Turkmenistan, both geographically as well as historically, is an extension of 
the Iranian plateau in the north, and the same features of aridity and settlements where 
water is found apply here. 



14 Chapter I 

The natural resources of Iran are many, but in antiquity few were exploited. In 
antiquity copper was mined, one might better say stripped from on, or just below, the 
surface, in the central deserts of Iran at Anarak and north of Kerman. Metallurgy 
developed in Iran early with alloys of copper, such as bronze, soon superceding 
copper. Tin was mined in Azerbaijan near Tabriz while lead, zinc, and iron were 
found in the mountains of Khurasan and Kerman. Gold and silver were also mined in 
antiquity, all of which made Iran a source of metals for craftsmen all over the Near 
East. Local craftsmen made Iran famous for its luxury arts and crafts throughout 
history, and textiles, rugs and ceramics were items of trade in the past as at present. 
Wood too was important for the crafts as well as for building, and one may assume 
that the mountain slopes were more covered with trees in antiquity than today, 
although no great changes in climate or water supply can be detected. 

The area of present day Iran is vast with many differences in climate and soil, from 
the jungles of Gilan to the lifeless mountains and deserts of Baluchistan. On the plateau 
the climate is continental with cold winters and hot summers, and settled life 
flourished in sheltered valleys where a water supply was available. Only on the edges 
of the plateau did this pattern vary. On the south Caspian coast and in Khuzistan 
ample rainfall enabled rice and other subtropical foodstuffs to grow. Yet the intense 
summer heat in Khuzistan makes any activity there virtually impossible for months 
of the year. The southern seacoast, extending to the Indus River, is too arid to produce 
anything except bare subsistence for a small population. On the plateau we find what 
one may describe as oasis settlements, either a series of small oases, or villages or urban 
centers with surroundings, such as Isfahan or Shiraz. The urban center may be more 
or less stationary because the water supply is constant, such as in the city of Hamadan. 
Or the city may shift its location over centuries, such as Nishapur. Also the names of 
cities appear and disappear or change. Sometimes the name of a town vanished while 
the district retained the name of the town, such as Qumis (Komisene) in Khurasan. 
Identifications of archaeological sites thus present problems, although, as noted above, 
the possibilities of ancient settlement were little different from what they are today. 
Nonetheless the very number of those possibilities in each valley or plain is not small, 
since present day villages were once large towns, and cities of today were once 
villages. Possibly nowhere on earth do we find evidence of the ebb and flow of 
history, the rise and fall of cities, with consequent deserted ruins, so prominent on the 
surface of the soil, as in Iran, Afghanistan and Central Asia. When the soil becomes too 
saline or the water table sinks, people will abandon a town or village, and the mud 
walls and ruins of houses will remain visible for years or even centuries. Sometimes 
entire cities, like Bam east of Kerman, were abandoned, providing tourist sites at the 
present time. Nowhere does the phrase — vicissitudes of time and the changes of 
fortune - seem more apt in its visible evidences than in this part of the world. 

AFGHANISTAN 

The extension of the Iranian plateau to the east ends in a great mass of mountains, 
which like the Zagros range in the west, has been a refuge area for many peoples. The 
geography of what is today Afghanistan and the western borderlands of Pakistan, for 
our purposes can be best understood as the area of four drainage systems where 



Geographical Survey 15 

agriculture and settled life for little more than a handful of people were possible. The 
mountains, from which all rivers take their sources, in the center of the country and in 
the east, have inhibited close contacts between the various parts of the country which, 
in any case, is a political creation of recent times. The Hari River with its principal 
city of Herat, loses its water in the sands now in the Turkmen SSR. Similar to the 
Khurasan province of Iran, historically Herat usually has been part of a western 
kingdom or empire, rather than of the east. On the other hand, Herat has served as a 
pivotal area for trade and invasion routes from the north to the southeast, to the plains 
of the Indian subcontinent, as much as a station on east-west trade routes. The valley of 
the Hari Rud is fertile, and it has been a source of wealth for the inhabitants enabling 
them to develop in the past a center of civilization in this region. The central 
mountains have been a barrier for communications between Herat and the Kabul 
valley in the east, which is revealed in culture by the Persian orientation of the former 
and the Indian for the latter. To the south, the road to Qandahar and thence to India 
crosses barren plains and rivers, but there is no formidable barrier to contacts, either 
between Herat and Seistan, or Herat and Qandahar. To the northeast of Herat, over a 
low range of mountains, lies the district of Badghis with the Murghab River, also 
draining north, but into the Merv oasis. This district, at present called Qala-ye No 
after the main town, is primarily an excellent pastureland with some fertile valleys, 
but which could never support a population comparable to the Hari Rud valley. If it 
were not for the summer wind of '120 days,' summers in Herat would be intolerable, 
since the low altitude of the city (c. 690 m.) would not provide relief from the hot 
sun. Other ancient places of settlement in the area included Fushang (today: 
Zendajan), some 45 km. to the west of Herat on the river, Aspozar or Isfizar (today: 
Shindand or Sabzevar) south of Herat, and Merv-e Rud (today: Bala Murghab) on 
the river of the same name. One may presume that ancient and modern centers of 
population differed little in their locations on this part of the Iranian plateau. 

The second system of drainage in Afghanistan empties the present Farah, Khash and 
Helmand Rivers, as well as smaller streams, into the perennial lakes or hamuns of 
Seistan. The Helmand is a 1300 km. long river, and in the spring it carries a large 
volume of water along its course. We have already mentioned Seistan in the 
discussion of Iranian provinces, and if the Afghan region is added, as it should be from 
an historical viewpoint, the entire area was very extensive and a center of civilization 
from ancient times. Seistan was larger than the comparable Merv and Bukhara oases 
in the north, and irrigation was widely practiced here in antiquity. The lower 
Helmand River basin was much more cultivated than it is today, but one must go up 
the river to the district of Qandahar and the confluences of the Arghandab, Tarnak 
and other rivers with the Helmand, to find comparable extensive cultivation and 
centers of population. On both sides of the lower course of the river are deserts, and 
there is no reason to suppose that in antiquity the cultivable land was much more 
extensive on either side of the river than it is today. The name Helmand, (Avestan: 
Haetumant) means 'rich in dams,' an apt description of a river which even today is 
the source of irrigation for a large expanse of land. The valleys of the many rivers and 
streams which descend southward from the central mountain massif are also fertile 
and productive, and one may suppose that this area was the homeland of the Pashtun 
or Pathan tribes to be discussed in the next chapter. The district of Qandahar, 



16 Chapter I 

however, not only maintained contacts with Seistan down the river, but also with 
Ghazna, 'the treasury' and Kabul. In antiquity the Qandahar-Ghazna route was on the 
border between Iranian- and Indian-speaking populations. 

Strictly speaking, the third drainage system of Afghanistan, the Kabul River and its 
tributaries, in ancient times was part of the Indian sphere of influence rather than the 
Iranian, and the river itself was part of the Indus River system. There are three large 
valleys which have been centers of settlement according to archaeologists, the 
Panjsher with the Koh-e Daman valley, the Kabul and the Logar valleys. The first 
valley is the highest in altitude (c. 2800 m.), and it lies north of the Helmand 
watershed as the first valley of the Kabul River basin. The Kabul valley (alt. c. 2200 
m.) was important from ancient times as the crossroads of routes from north and 
south, and on the edge of the descent to the plains of India. The mountains to the^east 
of Kabul quickly give way to lower altitudes, and the plain of Nangrahar (or 
Jalalabad) averages about 600 m., very hot in summer. To the north of Kabul the 
valley of Koh-e Daman is lower than Kabul, and with an average altitude of 1600 m., 
protected by the high Hindukush mountains to the north, the climate is also milder. 
Other, narrower valleys, such as the Ghorband and the high Bamiyan valley (c. 2800 
m.), were less populous in antiquity than Koh-e Daman. The ethnic populations of 
these areas will be discussed in the next chapter, but suffice it to say that in all of them 
mixtures of Iranian and of Indian speakers occurred in ancient times. The high 
mountains of the Hindukush and the narrow valleys of Nuristan (formerly 
Kafinstan), have created many refuge areas where people have fled from invasion or 
persecution, a paradise for linguists and anthropologists. Mining was practiced in the 
Hindukush mountains in antiquity, and the principal source of lapis lazuli was in 
Badakhshan on the northern slopes of the mountains. 

The northern slopes of the Hindukush, which is the watershed between the Kabul 
and Oxus River basins, have many fertile but narrow valleys, through which streams 
descend to the Oxus or Amu Darya. But it is the plains of Turkestan which were 
covered with many irrigation canals in older times, and which provided cultivation 
adequate to support a large population. 

The altitude of the plain of Turkestan varies from 350 to 400 m., a considerable 
drop from the mountains to the south, and it is in these plains that ancient cities 
existed where today their counterparts of Sheberghan, Aqcha, Mazar-e Sharif and 
Kunduz flourish. The largest city on the plains was Balkh, called 'mother of cities' by 
the invading Arabs, but today it has been replaced by Mazar-e Sharif. Many streams 
from the mountains do not reach the Oxus but are either used for irrigation or simply 
evaporate and vanish. The land on the southern banks of the Oxus on the whole is flat 
and sandy, while on the northern, present Soviet side, cliffs and mountains 
predominate. To the west, the plains of Turkestan become the Kara Kum desert, 
while to the east the mountains of Badakhshan lead to Wakhan and the Pamirs, where 
only hardy nomads can eke out an existence. 

The central mountains of Afghanistan, where the four river systems take their rise, 
have never been centers of population, and only the Bamiyan valley has had sizable 
permanent settlements in the past, for it was situated on a route leading from Balkh 
over the mountains to Kabul. In summary, one may conclude that settlement patterns 
in the area of present-day Afghanistan were concentrated in the hills and plains 



Geographical Survey 11 

around the central mountains, which were well watered, such as Seistan, Herat, 
Qandahar and the plains of Turkestan, except for the mountain valleys around Kabul, 
which combined provided sustenance for a large population, hence a base for power 
for a kingdom. The mountains to the east are parallel to the Zagros mountain range 
on the western side of the plateau. In both many narrow valleys and series of 
mountain ranges have provided regions of retreat or refuge for various peoples, and 
both are the farthest extent of Iranian-speaking peoples, to the lowlands of India in the 
east and to Mesopotamia in the west. Like the Zagros, the eastern mountains of 
Afghanistan were formerly more wooded, but the lack of rainfall and the extremes of 
temperature with a severe continental climate all over the plateau, never permitted 
the growth of population similar to the Indus, or Tigris-Euphrates lowlands. The 
people of the lowlands were rarely tempted to venture far into the mountains, 
whereas the rich, alluvial plains were always an attraction for the hardy inhabitants of 
the highlands. The relative inaccessibility of the sea coasts from the plateau did not 
induce the Iranian peoples to become seafarers, and neither the Caspian nor the 
Persian Gulf with the Arabian Sea played an important role in history for the peoples 
of the plateau, and Central Asia was even more isolated from the sea. 



CENTRAL ASIA 

A comparison of the altitudes of the principal cities at once reveals that Turkestan or 
Central Asia lies off the Iranian plateau. In the west the Caspian Sea depression extends 
across the Kara Kum and Kizil Kum deserts, which are really one desert separated by 
the Oxus River, to the mountains which descend from the Pamirs and the T'ien Shan 
range to the northeast. Ashkabad, capital of the Turkmen republic and near ancient 
Nisa, is 240 m.; Merv is slightly higher while Khiva in the delta of the Oxus is less 
than 1 00 m., and Bukhara is similar to Merv. To the east altitudes rise and Samarqand 
is almost 900 m. in some areas of the extended city, but Tashkent is only 455 m. An 
extension of lowlands to the east lies between mountains in the fertile Ferghana valley 
where Leninabad (former Khodjent) is 400 m., Kokand 396 m., Andijan c. 450 m., 
and Margilan 576 m. This valley provides the easiest access from the west over the 
mountains to Chinese Turkestan and the city of Kashgar. Just as the mountains to the 
south of the Oxus have many valleys descending to the river, so do the mountains of 
present-day Tajikistan to the north. The Hissar-Alai mountain range has three main 
rivers which water narrow but fertile valleys, the Surkhan Darya, the Kafirnigan and 
the Vakhsh. The last is the longest and takes its origin on the Alai plateau which 
provides a trade route to Chinese Turkestan, as well as excellent pasturage for the 
flocks and horses of nomads. In history the lands watered by the three rivers have 
been more closely tied to the plains of Afghan Turkestan than with Samarqand or 
Bukhara. The climate is also milder in winter here, protected by mountains to the 
north, than it is on the open areas to the west subject to winds blowing from Siberia. 
Dushanbe, the capital of Tajikistan, lies 824 m. high and has hot summers but mild 
winters. The great mountain complex of the Pamirs, with many glaciers, provides 
water for three great rivers, the Indus to the south, and the Oxus and Jaxartes (Syr 
Darya) to the west, not to mention streams flowing eastward into Chinese Turkestan. 



18 Chapter I 

The delta of the Oxus River was an important center of culture in the past 
(Khwarazm), but not the delta of the Jaxartes. 

Khwarazm (present Khorezm) is the low-lying land south of the Aral Sea mostly 
on the west side of the Oxus River. The eastern bank is more hilly than the opposite 
bank, and water for irrigation is taken from the river predominantly on the western 
side. Archaeologists have found ancient remains here, and one may infer that an early 
center of civilization flourished in this fertile region. The question of the ancient 
course of the Oxus River has never been conclusively resolved. For there are those 
who still claim that it emptied into the Caspian Sea at times, flowing into a depression 
known as Sarikamysh and then down a channel called the Uzboi, even though there is 
no evidence that the Oxus did flow into the Caspian Sea in ancient times. The situation 
in prehistoric times is unknown, and accounts of the river changing its lower coutse in 
the Islamic period are uncertain. In any case, the center of Khwarazm was located on 
the river over a hundred kilometers from the Aral Sea, below any turning off to the 
Caspian Sea, which appears unlikely in the historical period with which we are 
concerned. 

The antiquity of settlement in Khwarazm is matched by the Merv oasis, with the 
added advantage of lying on the important trade route from Iran to Central Asia and 
China. Its location seems to have made this oasis on the Murghab River more 
important than the lower course of the Hari Rud (called the Tejend), which also lost 
its water in the desert, although the town of Sarakhs on the latter was similar to Merv 
on the Murghab. Water being the key to life in the deserts of Central Asia, river 
valleys are natural places for settlements, but the rise of towns on them is usually 
determined by their location on trade routes. Merv, situated in the Kara Kum desert, 
was the stopping place for caravans plying between the Iranian plateau and Bukhara 
and the Zarafshan valley. 

The oasis of Bukhara, like that of Merv, utilized the water of a river, here the 
Zarafshan, before it disappeared in the sands, and like Merv, the rise of Bukhara must 
have followed the development of trade. The prosperity of the Merv oasis, the 
Bukhara oasis, and many other sites in Central Asia, depended on the efficacy of the 
irrigation systems. In fact, so important is this all over this area, as well as Iran and 
Afghanistan, that the ancient Iranians may be called the practitioners of an irrigation 
civilization par excellence. It was extensive development and expansion of irrigation 
systems which brought prosperity and civilization to the settlements such as Merv, 
Bukhara and Samarqand, although all three owed their rise to their locations on trade 
routes. 

One of the ancient centers of settlement in Central Asia was the Ferghana valley, a 
rich area enclosed by mountains to the north, east and south, some 300 km. long and 
70 km. wide, watered by the Jaxartes. To the west is a narrow desert strip called the 
Hungry Steppe,' some 7 km. wide, which continues into the larger Kizil Kum desert; 
so the Ferghana valley is virtually an isolated valley varying from 350 to 900 m. in 
altitude. The relatively low altitude, as well as the mountains, protect the valley from 
the cold north winds of winter to which Tashkent and Bukhara are subjected. The 
middle of the valley has much barren steppe land, but the foothills of the mountains 
are relatively rich in verdure. Archaeologists have reavealed very early settlements in 
the valley, and the existence of various minerals in the surrounding mountains 



Geographical Survey i9 

undoubtedly further enhanced the attractions of the irrigated land as a good place to 
settle, for mining was also practiced at an early date. 

To the north of the Ferghana valley, and in the foothills to the east of the Kizil Kum 
desert was the oasis of Chach, today Tashkent, which was also settled in early times, 
although probably not as extensively as either the Ferghana valley or Samarqand. To 
the east of Chach the mountains hindered both communications and settlement such 
that the main road to the east ran north skirting the mountains, and then turned east 
over low passes to the Talas River valley. Again skirting the Kara-tau and Ala-tau 
mountain ranges, the road continued eastward towards the Altai Mountains. The vast 
steppes of northern Central Asia and southern Siberia were traversed by Iranian 
nomads called by a generic term, Sakas, when we first hear of them in the sixth 
century B.C. But here too, as in the south, the valleys of the Chu and Hi Rivers had 
settlements from early times. 

Not only did Iranian nomads reach the Altai mountains, but they also spilled into 
Chinese Turkestan, which, before the Turkic expansion, was in great measure 
Iranized. Today only some 20,000 Iranian speakers survive in the Sarykol valley of the 
Chinese Pamirs with the principal town of Tashkurghan. The northern and southern 
sides of the Taklamakan desert in Chinese Turkestan were inhabitated by Iranians, as 
well as others, and traces of ancient settlements have been found on dried-up stream 
beds far into the desert. The extreme aridity of the entire area makes any settlement 
entirely dependent on irrigation, and in the southern rim of the Tarim Basin, as it is 
called, the only oasis of importance is Khotan because of two rivers which bring 
water down from the Kunlun mountains to the south. The Kunlun mountains and 
the Tibetan plateau made communications with the south almost impossible, and 
only two arduous routes to the west were important in history. The first, and most 
used, northern route we have briefly mentioned above. It followed the Surkhab 
(Turki: Kizilsu) River through the Alai plateau over the mountains to Kashgar. The 
southern route followed the Wakhan corridor in present Afghanistan through the 
eastern Pamirs to the valley of Sarykol. The easiest access to the Tarim basin of 
Chinese Turkestan was from the north over several passes of the Tien Shan 
mountains. The eastern part of the Tarim, on the borders of China, comprises a salt 
desert, called Lop Nor, once a prehistoric sea, stretching almost 300 km. in length. 
Human habitation was always at a minimum here. On the northern side of the Tarim, 
however, several oases have flourished in history, such as Aqsu, Kucha and Turfan, 
probably because water is more available from streams descending from mountains 
to the north, and there is more fertile ground, not as eroded as the characteristic loess 
earth of the south. 

The reason why Chinese Turkestan has been included in the geographical survey of 
ancient Iranian civilizations is that Iranian Sakas lived in the oasis of Khotan and 
vicinity for centuries before becoming absorbed by Turks, and the northern string of 
oases formed a trade route to China over which Sogdian merchants travelled and 
made settlements. The oasis of Turfan has yielded valuable documents in the Sogdian, 
Parthian and Middle Persian languages, of inestimable value in many fields. We will 
return to this in the chapter on demography. 

This survey of the lands where Iranians lived in the past indicates the common 
features of terrain in Iran, Afghanistan and Central Asia. We must repeat that 



20 Chapter I 

mountains, deserts and oases, with settlements based on irrigation, are the hallmarks 
of life in this part of the world. There was rarely any excess of agricultural produce, or 
rather foodstuffs, to trade; rather minerals and objects of craftsmanship, such as 
textiles and carpets, were the mainstay of trade from ancient times. Also the difference 
between the dweller in an oasis, and one on the bleak steppes, mountains or deserts, 
was always marked. The contrast also between the verdure of a well-watered oasis, 
which seemed a haven of security, and the forbidding land outside, was sharp, and this 
conditioned human attitudes about nature and the land in which the Iranian lived. 
Extremes of temperature and frequent earthquakes, flash floods, and other ravages of 
nature, tested the endurance of humans, and left them ever ready to migrate to the 
more fertile and less harsh lowlands of India or Mesopotamia. If anything has 
conditioned the lives of Iranians, it is this sharp contrast in nature, between the huge, 
arid, barren, forbidding mountains and deserts and the irrigated garden surrounded 
by mud walls to keep out the sand and the encroachments of a hostile outside world. 
The word 'paradise' is Iranian in origin, and it originally meant an enclosed garden 
where the ruler hunted his favorite game which had been introduced into the 
enclosure for that purpose. 

From this geographical survey of Iran, Afghanistan and Central Asia, it is clear that 
there could be neither unity nor large populations here like the Nile and Tigris- 
Euphrates River basins. Vast areas, for the most part, could only be ruled for a short 
time by either nomadic or tribal states, or by some kind of feudal alliances. The 
problems of communication and holding allegiances were enormous, yet some sort of 
adherence or allegiance to Iranian culture permeated the entire area, at least until the 
expansion of the Turks changed much of the demography. The conflict between 
steppe and sown area also has been a thread running through the history of this part of 
the world. In general one may say that geography and natural conditions have always 
played an important role in this vast area, and nature has not been predictable or 
gentle. 

The earthquakes and other natural catastrophes are constant features of this entire 
area, so man has lived in the shadow of an unpredictable nature over millennia of 
years. For the history of culture and civilization, however, the means by which people 
overcame their isolation and traded with each other, exchanging ideas as well as 
articles of trade, was of primary importance. River valleys, passes over mountains and 
watering places in the desert assumed a greater importance than in other regions of 
the globe. This is why an understanding of geography, which might not be so 
relevant to the study of the history of others, in the case of the lands occupied by 
Iranians is of paramount importance, and in the pages to follow constant reference 
will be made to the geography of the area. 



CHAPTER II 
DEMOGRAPHY 



Anthropologists have made many studies of towns and villages in Iran, Afghanistan 
and Soviet Central Asia. Linguists have investigated varieties of dialects, but ancient 
historians have been overwhelmingly concerned, on the whole, with events reported 
by chronicles or other written sources, perforce mainly concerned with the royal 
courts or more rarely with religious leaders or intellectuals including poets and men 
of letters. We should remember that few written records of the ancient past are 
concerned with the life of the common folk, and even among them, nomads or 
villagers are rarely mentioned, civilization and writing being bound to the urban 
environment. Nonetheless, populations did change with rather important conse- 
quences for the history of the entire area. One may divide people in the Iranian 
cultural area between settled people, nomads, and mountaineers, plus urban dwellers, 
although other patterns of classification might be made. The modes of life of these 
four divisions have been constant throughout history although the people who 
followed the various ways of life have not remained static. To understand the past of 
the common folk we must consider the present and then frequently interpolate from 
know to unknown. That is why anthropological and even sociological studies, as well 
as linguistic research, must be heeded by anyone trying to recover the past of this part 
of the world. A survey of relevant bibliography shows many gaps. 

Literature : Iran, General: There is no comprehensive survey of the peoples of Iran except in general books 
on the country which include chapters on demography. Usually these accounts are brief and inadequate, 
but one should be mentioned, since it is more comprehensive than others: N. Y. Kislyakov, ed., Narody 
Perednei Azii (Moscow, 1957), 173-308, dealing with the non-Turkic population of Iran. Individual 
studies of both settled and nomadic peoples exist; two brief chapters, on the 'people' (318-55) and 
'distribution of the population' (356-93) are relevant to our concern in Persia, Geographical Handbook 
Series, Naval Intelligence Division (London, 1945). Few of them have any relevance to pre-Islamic times; 
some, however, do provide interesting information which can be interpolated back to the past, for 
example the census. The Iranian Statistical Centre in Tehran is in charge of the results of the census; the 
first real census was made in 1956 and published in 1 10 parts in Tehran beginning in 1961, while the next 
census, of 1966, is still being printed. Several general studies on the population of Iran have appeared; F. 
Firoozi, "Demographic Review, Iranian Censuses," The Middle East Journal, 24 (Washington, 1970), 220- 
28, and a popular discussion by H. Goblot, "La structure de la population de l'lran," L'Ethnographie, 57 
(Paris, 1964), 33-54. Before the 1956 census most statistics on population were mere guesses. While 
reliable figures are lacking the present locations of settlements and cultivation can provide clues to past 
settlement and population patterns, for the land and availability of water have not changed much over the 
centuries. 

It is not germane to our task to list studies of physical types or races in Iran, since mixtures of cranial 
types and skeletons are found in the earliest excavations, and only extensive comparison of these with 
modern skulls might give us an idea of significant racial changes in Iran's history. Very roughly, one 
might say that today in the northern part of Iran brachycephals with curved 'Armenoid' noses are more 
common than in the south where narrow-faced dolichocephalics may predominate, and this seems to have 
been true of the past as of the present. On the physical anthropology of Iran at present see H. Field, 
Contributions to the Anthropology of Iran, Field Museum of Natural History, vols. 29-30 (Chicago, 1939; 
reprinted 1968). Most publications of archaeological excavations in Iran have a section on skeletal 



22 Chapter II 

material, and one general survey of ancient man, based on remains from Tepe Hissar is by W. M. 
Krogman, "The peoples of early Iran and their ethnic affiliations," American Journal of Physical 
Anthropology, 26 (Philadelphia 1940), 269-308. For earlier man see C. S. Coon, "Iran," in H. Vallois, 
Catalogue des hommesfossiles (Macon, 1953), 267-70 and M. Cappien, The Iranians of the Copper/Bronze 
Ages, H Field Research Projects (Miami, 1973), 60 pp. The only discussion of peoples in western Iran in 
early historic times is the study by I. M. Dyakonov, "Narody drevnei prednei Azii." in Predneaziatskii 
Etnograficheskii Sbornik, 1, Trudy Institut Etnografii, 39 (Moscow, 1958), 18-39. 

Many publications exist on various tribes or small groups living in Iran, but the majority of settled 
population has received little attention insofar as questions of ethnogeny are concerned. The main Iranian 
tribal people are the Kurds, and several books discuss their origins and early history, such as it is known, as 
B. Nikitine, Les Kurdes (Paris, 1956) and esp. O. Vilchevskii, Kurdy, Trudy Instituta Etnografii, 67 
(Moscow, 1961), 165 pp., which is concerned with the origin of the Kurds. A special sect of the Kurds is 
the subject of S. S. Ahmed, The Yazidis, their life and beliefs, H. Field Research Projects, no. 97 (Miami, 
1975), 485 pp. A large bibliography on the Kurds is useful also for ancient questions: S. van Rooy, 
International Society of Kurdistan Bibliography, 2 vols. (Amsterdam, 1968). See also N. A. Aleksanian, 
Bibliografiya Sovetskoi Kurdskoi Knigi 1921-1960 (Erevan, 1962). The Lurs to the south of the 
Kermanshah-Hamadan road, are not as well studied, as the Kurds, but we have some information on them 
in Persian Publications: 'Ali Muh. Sakl, Luristan (Khurramabad, 1343/1969), and H. Iz5dparrah,/ir/j<}r-e 
bastani ve tarlkhl Luristan, 2 vols. (Tehran, 1350/1972 and 1353/1975), concerned with the antiquities and 
history of Luristan as well as the present population. Anthropological articles and monographs exist on 
the Kurds and Lurs, which may be found in the bibliographies above. 

The peoples of the Caspian Sea coast, Talysh, Gilan and Mazandaran, had and have dialects and 
characteristics different from the rest of Iran. On the people of Mazandaran consult a series of monographs 
in Persian on villages and districts from east to west, beginning with Alasht by HQshang Por KarTm, 
published by the Ministry of Culture and Art (n.d., c. 1 976), Yush by Situs Tshbsz (University of Tehran, 
1342/1964), Orazon byjalal Al-Ahmad (Tehran, 1954), and Feshendak by H. Pur KarTm (University of 
Tehran, 1341/1971). All of these deal with anthropological, linguistic, folkloric and other matters. For 
material remains along the entire Caspian coast we have a series of volumes by ManOchehr SotDdeh, Az 
Astara ta Astarabad (Tehran, 1 349/1 971 , and foil.). Other references may be found in works on dialects of 
this region, on which see below. The Encyclopaedia of Islam (old and new editions) has information and 
bibliography on individuals, places and tribes. 

The southern tribes of Iran, which are not Turkic, have been studied by anthropologists presenting 
material of interest for an historian; cf. D. Ehmann, Balftiyaren (Wiesbaden, 1975), and V. V. Trubetskoi, 
"Bakhtiary," in M. S. Ivanov, ed., Etnicheskie protsessy i sostav naseleniyakh v stranakh Prednei Azii 
(Moscow, 1 963), 1 41 -72. Further see F. Barth, Nomads of South Persia, The Basseri Tribe (Oslo, 1 961 ) ; V. 
Monteil, Les tribus du Fars (Paris, 1966); and M. S. Ivanov, Plemena Farsa, Trudy Instituta Etnografii, 43 
(Moscow, 1961); R. Loffler, "Die materielle Kultur von Boir Ahmad, Slidiran," Archivfiir Velkerkunde, 
28 (Wien, 1974), 61-1 42; and in Persian, ManOchehr Lam a, Farhang-e 'amyaneh-e 'ashalr Buyer Ahmedl 
ve Kohgiltiyeh (Tehran, 1 349/1 971 ). On the primitive people of Bashakird, to the west of Baluchistan, see 
I. Gershevitch, "Travels in Bashkardia,"JJMS, 46 (1959), 216-24. A mountain group in southwest Iran 
are Les Papis by C G. Feilberg (Copenhagen, 1952). Other references may be found in the works listed. 

To the southeast, the chief group of Iranian nomads are the Baluch who extend to the Indus River. On 
them see M. G. Pikuhn, Beludzhi (Moscow, 1959), and on the northernmost group E. G. Gafferberg, 
Beludzhi Turkmenskoi SSR (Leningrad, 1 969). Many publications from Pakistan exist on the Baluch, such 
as Muh Sardar Khan Baluch, History of the Baluch Race and Baluchistan (Karachi, 1958), but most are of 
little value for our purposes. A non-Iranian people, and possibly among their predecessors in Baluchistan, 
are the Brahuis, a Dravidian-speaking folk, a few villages of whom exist today in Iranian Seistan ; see G. H. 
Srasser, "Brahui in Persien," Bulletin of International Committee on urgent Anthropological and Ethnological 
Research, 2 (Vienna, 1959), 97-98. 

Perhaps the most important element for differentiating people, language, fortunately needs few 
references, since surveys and bibliographies exist on all of the Iranian languages; see "Iranistik, erster 
Abschnitt, Linguistik," of the Handbuch der Orientalistik ed. by B. Spuler (Leiden, 1 958) ; further T. Sebeok, 
ed , Current Trends in Linguistics, 6 (The Hague, 1 970), 1-135, and the useful handbook by I. M. Oranskii, 
Vvedenie v iranskuyu filologiyu (Moscow, 1960), trans, into French by J. Blau, Les langues iraniennes (Paris, 
1977). Needless to say, much information on ancient languages can be found in modern dialects, not to 
mention relatively new fields such as glottochronology or areal linguistics. 

Finally, a bibliographical note on handicrafts and technology and plants may be useful. Fortunately a 
handbook by H. E. Wulff, The Traditional Crafts of Persia (M.I.T. Press, Cambridge, Mass., 1966), also has 



Demography 23 

a large bibliography of relevant literature. The first chapter is on metallurgy in ancient Persia, on which 
cf. T. Wertime, "Man's First Encounters with Metallurgy," Science, 146, no. 3649 (1964), 1257-67. The 
classic works by B. Laufer, Sino-lratiica, Field Museum of Natural History publication 201, vol. 15 
(Chicago, 1919), discusses plant borrowings, and this is supplemented by E. H. Schaefer, The Golden 
Peaches of Samarkand (Berkeley, 1 962). Dictionaries of plants and trees, animals and minerals, in Persian or 
Arabic, contain archaic words and information about them. The Kitab al-jamahtr (Hyderabad, 1355/ 
1936) of al-BlrOnT, is perhaps the most famous book on minerals. While such works may seem to have 
little connection with ancient Iran, they are similar to Pliny's natural history, which contains items of 
historical significance. It must be remembered that unlike Greek or Roman history with a plethora of 
sources, we must use every piece of information in the little known east to attempt to reconstruct the 
scantiest of history. 

Literature: Afghanistan, general: Studies on the people of Afghanistan and northwest Pakistan are mostly 
concerned with linguistic minorities such as the peoples of the Pamir or Hindukush mountains, especially 
the Kafirs or Nuristanis. Several bibliographies are of great aid to the researcher which have been listed 
already in the first chapter on geography. Several general handbooks, including detailed chapters on the 
population, languages, etc., are W. Kraus, ed., Afghanistan, Erdmann, La'ndermonographien, 3 (Tubingen, 
1 972) ; D. N. Wilber, ed., Afghanistan (New Haven, 1 962), and L. Dupree, Afghanistan (Princeton, 1 973) ; 
each of which has a different viewpoint. Old works such as H. W. Bellew, The Races' of Afghanistan 
(Calcutta, 1880), may contain interesting items, but on the whole they are out of date. A special 
bibliography exists on Nuristan by S.Jones, An Annotated Bibliography of Nuristan (Kafiristan) and the 
Kalash Kafirs of Chilral (Copenhagen, 1966). On the Kafirs, see the work of P. Snoy, Die Kafiren 
(Universita't Frankfurt M., 1962), and further his "Nuristan und Mungan," Tribus, Nr. 14 (Stuttgart, 
1965), 101-48. In the same issue, pp. 11-99, consult F. Kussmaul, "Badaxsan und seine Tagiken." Studies 
of various languages in Afghanistan are listed in the same works, noted above, for the languages of Iran. 
Many scattered articles on various aspects of Afghanistan's ethnography, as well as history, numismatics, 
and others, are photographically reproduced in the journal Afghanistan, published by the Historical 
Society in Kabul. On physical types see G. F. Debets, The Physical Anthropology of Afghanistan, trans, by E. 
Prostov (Peabody Museum, Cambridge, Mass., 1970). Since Afghanistan has not had a proper census 
much work remains to be done in demography, as well as ethnography, cf, however, L. Dupree, 
"Settlement and Migration Patterns in Afghanistan", Modern Asian Studies, 9 (1975), 397-413. On the 
nomads of Afghanistan see C. Jentsch, Das Nomadentum in Afghanistan, Afghanische Studien 9 
(Meisenheim, 1973), also B. Glatzer, Nomaden von Charjistan, Beitra'ge zur Stidasienforschung, 
(Heidelberg, 1977). Other studies, on Turkic or Mongolian peoples of Afghanistan are not germane to 
our task here. 

Literature: Central Asia, general : Many Soviet publications have been devoted to the ethnogeny of the 
peoples of Central Asia and the Caucasus, and it is not possible to mention more than the most helpful 
general studies, many of which contain detailed bibliographies. Consult H. Field, Bibliography of Soviet 
Archaeology and Physical Anthropology (Miami, 1972). The series Narody mira, etnograficheskie ocherki, ed. 
by S. P. Tolstov, 2 vols. (Moscow, 1962-63) is especially useful for Central Asia and for the Caucasus, 2 
vols. (Moscow, 1961-62). M. S. Andreev, N. S. Kislyakov and others were especially interested in the 
ethnography of the Tajiks and wrote several articles about them; cf. vol. 1 of Narody Srednei Azii 
(Moscow, 1 962), 701-3. On the mountaineers of Tajikstan see O. E. Agakhanyants, Bibliografiya Patnira, 1 
(Dushanbe, 1968). Such works as Tajiki Karategina i Darvaza by N. Y. Kislyakov and A. K. Pisarchik 
(Dushanbe, 1976) contain much interesting material for historians. Consult also the various publications 
of the USSR Academy of Sciences such as Sovetskaya Etnografiya and Trudy Instituta Etnografii, especially 
the Sredneaziatskii Etnograficheskii Sbornik II, no. 47 (Moscow, 1959) where noteworthy is the article by 
L. F. Monogarova, "Materialy po etnografii Yazgulemstev," 3-94. Further see H. Field, Contributions to 
the anthropology of the Soviet Union (Smithsonian miscellaneous collections, vol. 1 10, no. 13, Washington, 
1948), and his Contributions to the physical anthropology of Central Asia and the Caucasus (Peabody 
Museum, Cambridge, Mass., 1968). 

The areas now occupied by Turkic-speaking peoples, Turkmenistan, Uzbekistan, Kazakhastan, and 
Kirghizistan, were once peopled by Iranians, so we are more concerned with archaeological remains of 
the ancient inhabitants than the present population. The results of archaeological expeditions are especially 
valuable, and the Khwarazmian expedition led by S. P. Tolstov plus the Southern Turkmen complex 
expedition, begun by M. E. Masson, have been especially noteworthy. See, for example, T. A. Trofimova, 
Drevnee naseleniye Khorezma, po dannym paleoantropologii (Moscow, 1959). 



24 Chapter II 

THE ENTIRE AREA 

All of the people of present Iran, Afghanistan and Central Asia can be simply divided 
into settled people and nomads, with various degrees of settlement and nomadism- 
pastoralism represented. Since the geographical features of this vast area, including the 
availability of water, have helped to determine in no small measure the population 
distribution in historical times as well as the present, the life-styles of the present 
people can throw light on the past. On the other hand, in any particular region it is 
frequently not possible to ascertain whether the population in the past was definitely 
nomadic or settled, or predominantly one or the other. The previously held 
supposition of some anthropologists that early man was first a hunter, then a 
herdsman and finally a settled farmer in a temporal progression, is not accepted today. 
The change from a pastoral to an agricultural use of land, or the reverse, can be 
observed in recent times, and must have been true in antiquity. The falling of a water 
table or salinization of farm land causes the abandonment of a village, while the 
persistence of a few years of drought in one area can change a flourishing oasis into a 
desert. Therefore, in regions which did not have a regular and abundant supply of 
water, changes in ecology must have occurred many times in the past. Certainly one 
of the prime tasks of governments in this part of the world has been to organize, 
regulate and promote the use and distribution of water. Of the three requirements, 
land, water and manpower, necessary to insure any flourishing culture on the Iranian 
plateau, water has been the most difficult to supply, and this has been true from 
antiquity. 

Throughout history the stronger have displaced the weaker in possession of good, 
watered land, and the weaker people either coexisted with their masters, or they 
moved to less desirable 'refuge' areas in the mountains or deserts. Settled oases, which 
were and are the dominant places of settlement over most of the plateau, presumably 
were early mixed in population. From the dawn of history we find a differentiation of 
labor in settlements, and one of the clearest groups, distinct or apart from the rest of 
the populace, were the smiths or metal workers. They had probably evolved from 
potters, who found that high temperature fires made glaze on pottery, and then 
experimented with other media. In any case, a caste of smiths seems to have developed 
in the Near East who were different from the local populations. Otherwise, 
specialization of functions or labor does not seem to have produced separate castes or 
the like. We may presume that people in the ancient Near East were separated by 
different languages or ways of life rather than by profession. Since nomads or 
mountaineers have not left records one must project backwards from known 
situations of later times. For example, much of the land on the plateau, other than 
barren deserts or mountains, is not fit for agriculture but only for grazing. So it was 
also in the past, and the conflict between the steppe and the sown area dates from the 
unrecorded past. Since not only history but writing itself is a product of settled life, 
the pastoralists remain on the dim fringes of settlements, little recorded and hardly 
appreciated. Consequently, our picture of the past is one-sided in many respects, and 
one can only note the continued existence of nomads and pastoralists, who usually 
appear in records only when they mount massive raids on the settlements which 
suffer from their depredations. 



Demography 25 

The routes of trade or the movement of peoples have been exceedingly constant 
throughout the history of Iran because of the dictates of geography and water. Thus, 
the easiest route of access from the Mesopotamian lowlands to the plateau has been the 
present route of Khanaqin, Kermanshah, Hamadan to Tehran, and then east to 
Khurasan. There are other routes over the Zagros Mountains from the west, but they 
are more arduous and were much less used than the principal route through the 
Kermanshah area. The Assyrians and succeeding conquerors saw the need to control 
this route as a key to rule over any pretended empire. A map showing the routes, 
which have followed essentially the same passes through the mountains, shows the 
primacy of the Kermanshah route, which is also well watered. One may presume that 
along this route, as well as others, settlements arose which began to trade with each 
other. At all times, of course, minor variations in settlements, as in the routes 
themselves, have taken place, but the main lines of communication have remained 
constant. 

Another feature of the plateau, which should be mentioned, is the great variety of 
fruits and agricultural products which grow in relatively small geographical areas. 
Because of the altitude, it is possible in many places to find apples, pears, and other 
fruits of a temperate zone in the mountain valleys, while a few kilometers distant, in 
an oasis protected from winds and cold, date palms and citrus fruits will flourish. This 
variety was always a feature of some mountain areas of Iran, and it in a sense parallels 
the people who may be very different from each other yet living only a short distance 
apart. Similar conditions obtain in Afghanistan and in Central Asia, but on the whole 
not so variegated as in Iran. Although local specialities of crafts as well as produce 
distinguish one area from another, the unity of the entire Iranian area in modes of life 
and responses to nature appears throughout history. Rather, the unity is one of similar 
diversity in response to the changes in landscape and availability of water and other 
natural phenomena. This is best observed in the different linguistic groups which 
exist later, especially before the expansion of Turkic speakers, beginning in the tenth 
century of our era. Even though many people did adopt new languages, the basic 
population continued to be the same though mixed with newcomers. There is no 
better way to distinguish people than by the languages they speak, but one must not 
neglect the settled-pastoral differentiation, and both are followed here in a survey of 
the entire Iranian area. 

IRAN 

Fars exemplifies the diversity of nomad and peasant, mountain and desert and other 
contrasts, perhaps better than any other province of Iran. From the present population 
of the province one must exclude Arabs who came after the Islamic conquest and the 
Turks, such as the Qashghai nomads. Today the Arab-speaking population along the 
shores of the Persian Gulf is sparse, and in antiquity the numbers must have been less. 
There has always been, it seems, contact across the Gulf, so some mixtures of 
populations would be expected. Since Elamitc clay tablets have been found in the 
region of present Bushire, it would be natural to assume the existence of early Elamite 
settlements along the coast. Unfortunately, we know too little about the Elamite 
language to even speculate about borrowings or influences from Elamite on certain 
dialects of Fars as opposed to others. We may suggest that the population of the coast 



26 Chapter II 

was engaged in fishing as well as transit trade and commerce with the interior, while 
the date palm, as across the Gulf, was the local staff of life for them. Since the coastal 
plain is narrow and the mountains behind it are barren and lacking in verdure and 
water, this area never enticed large nomadic tribes to spend winters here with their 
flocks. The coastal settlements were usually dominated by governments on the 
plateau, and the poverty of the coastal people kept them from playing a significant 
role in history. 

There are many ancient remains on the next step up onto the plateau to Kazerun, 
evidence of a flourishing area as early as Elamite times. The Elamites probably 
occupied most of the valleys and plains which extend from the northwest to the 
southeast, and it would be natural to assume that ancient pastoralists at times followed 
these valleys down to the rich alluvial plains of Mesopotamia for the winter. Much 
has been written on the far-flung ancient trade connections, which brought, for 
example, lapis lazuli from the mountains of Afghanistan as far as Egypt in early times. 
This has been adduced as evidence for very distant contacts between traders, but it is 
misleading. True, some hardy traders in antiquity may have traveled enormous 
distances like Marco Polo in later times, but they like he, were rare exceptions. Most 
trade was usually a process of exchange from one local group to another rather than a 
direct long distance contact between the consumer and producer. Wares usually were 
passed from one merchant to another, although sometimes they did travel long 
distances. Likewise, the presumed migrations of peoples over vast distances were 
probably even rarer than long voyages by merchants, and changes in material culture 
on the plateau are much more likely the result of diffusionary influences or even 
internal stylistic development, rather than the intrusion of a new population. Great 
movements of peoples like the Arabs and Turks in more recent times or the Iranians 
onto the plateau were probably just as few in prehistoric as in historic times, and one 
must be careful about the attribution of a change in pottery designs to a mass invasion. 
Also the movements of people noted above were the migrations of nomads or 
pastoralists, who were rarely great originators of complex, new cultures as compared 
to the settled folk. 

The upper steps of the Zagros range, known today as the sardslr 'the cold zone' were 
eminently suited for pastoralists, and we may assume that much of Fars, such as the 
present areas of Pasargadae and Abadeh were very sparsely settled before the coming 
of the Iranians. It is not the purpose of the present volume to discuss archaeology, 
which is covered in Barthel Hrouda's Vorderasien, Handbuch der Archa'ologie, 
(Mtinchen, 1971), but since almost all of our evidence of pre-Archaemenian Iran is 
from archaeology, it cannot be ignored in presenting a background for the historical 
periods. The material remains in prehistoric southern Iran point to an Elamite 
occupation of much of the area. It is, of course, impossible to determine whether the 
people of the mountainous areas of Fars were the same as those in the valleys and 
plains, but the extensive survey of the Marvdasht plain, where Persepolis is located, 
by William Sumner shows a unity in the extensive occupation of all parts of the plains 
from very early times. 1 From surveys, based primarily on surface potsherd finds, 

1 W. Sumner, Cultural Development in the Kur the Persepolis Plain and Shiraz Area," Iran, 6 
River Basin (Ann Arbor, Michigan 1972), Univer- (1968), 168-70. 
sity Microfilms. Also see P. Gotch, "A Survey of 



Demography 27 

similar occupation periods have been found elsewhere in Fars, especially in those oases 
where one would expect a continuous occupation, such as Fasa and Darab. 2 
Unfortunately some areas of Fars province archaeologically are virtually terra 
incognita, such as the districts of Lar-Khunj and Ivaz in the south, and the 
Farrashband-Kazerun valley west of Shiraz. There is no reason, however, to assume 
that the population of these areas was different from that of the large Marvdasht plain. 
One may suppose that some tracts of lands which are occupied today or were in the 
early Islamic period, in antiquity were either empty of people or sparsely settled. 
Further, since trade was an important factor in the development of civilization, it is 
not fool-hardy to conjecture that the people of Fars early traded foodstuffs and animal 
products between the inhabitants of the cool and warm zones of the province, with 
little need or even incentive to go beyond the provincial borders. 

The distribution of Iranian dialects today in Fars may provide an indication of 
settlement patterns in antiquity. The most obvious and distinctive dialects are those 
spoken in the mountainous area of southern Fars, the dialects of Laristan. It is 
significant that many of these dialect speakers are today Sunni Muslims, separated 
from the Shiite majority of Persians. Religious difference surely contributed to the 
maintenance of these dialects as much as geographical isolation, but whether the 
difference between the people of Laristan and others in Fars is a phenomenon of 
Islamic times, or has antecedents in the pre-Islamic past, cannot be determined. 
Perhaps all that can be said is that given the geographical isolation of Laristan, an 
ancient differentiation of population in this part of the province, compared to other 
areas, would not be unexpected. Here mixtures of peoples, including some negroid 
strains, especially along the coast, took place in antiquity as well as Islamic times. It 
should be noted that the existence of an Iranian dialect spoken in Oman today, 
Kumzari, indicates the two-way traffic of peoples moving across the Gulf in the past. 3 

One would expect the Zagros mountain range to contain pockets of linguistic 
differences from standard Persian, but by now most of these have been lost, although 
it must be admitted that little investigation has been made of possible survivals. One 
of these is the dialect of Davan, a village in the mountains northeast of Kazerun. 4 This 
is not the place to go into the peculiarities of dialects, but historically one may propose 
different strata or waves of Iranian dialects which spread over Fars province, 
absorbing pre-Iranian tongues completely. 

The northwestern part of Fars province has rugged mountains and here live today 
Luri-speaking tribes, the Boyer Ahmad and Kohgiluye. Strictly speaking they are 
mountaineers, even though transhumance is practiced by many of them. The way of 
life of these contemporary people can have differed little from that of the their 
ancestors. 5 It is significant to note that Islamic geographers of the tenth century, 
writing in Arabic, mention the existence of the Khuzi language in these mountains, 
which has been interpreted as a continuation of an ancient Elamite dialect. 6 Today the 

2 P. de Miroschedji, "Poteries Elamites du Fars H. Mahamedi, "Three Iranian Dialects of Fars, 

Oriental," Bastan Chenassl va Honar-e Iran, 7-8 Davani, Avazi and Koroshi," Iranian Studies, 14 

(Tehran, 1971), 60-67. (1982). Note that the people of Davan are Shaikhis, 

B. Thomas, The Kumzari Dialect of the a special sect of Shiite Islam. 

Shiluh tribe," JRAS (1930). 5 See Loffler, supra, Boir Ahmad, 62-141. 

G. Morgenstierne, "Stray Notes on Persian 6 References in Schwarz, op. cil. [ch. 1, n. 14], 

Dialects, -NTS, 19 (Oslo, 1960), 123-29, and 406. 



28 Chapter II 

language of the Boyer Ahmad tribes can be called a Luri dialect, quite close to Persian, 
and thus this northern part of Fars province is really part of the mountainous area of 
Luristan. It is possible that the Lurs, like the people of Laristan in the south, and others 
in mountainous areas, represent the vestiges of the first wave of Iranians who invaded 
the plateau and whose dialects were replaced in the valleys and on the plains by New 
Persian. An indication of this hypothesis, but no more than that, is the common 
vocabulary of Luri and Laristan dialects, where certain common words and usages are 
not found in Persian. 7 The fact that New Persian is a wedge between these outlying 
dialects is best explained by the wave hypothesis, that an earlier substratum of Iranian 
dialects was gradually replaced by a standardized New Persian dialect, but the time 
when this occurred is unknown. In any case, we may assume a parallel earlier process 
whereby the original Iranians in Fars province gave their tongue to the 
autochthonous inhabitants in the first millennium B.C., whereas the mountain areas 
retained their original language and customs longer than the plains. 

The plains of Khuzistan from the earliest times were bound to the neighboring 
mountains of Luristan in a symbiosis of peoples. Not all of the population practiced a 
yearly migration to the mountains in the summer and to the flatlands in the winter, 
however, but enough of them did to mix the population such that Khuzistan was 
different in this respect from the lands to the west, in the basin of the Tigris and 
Euphrates. Contact between the peoples of Mesopotamia and Khuzistan was much 
easier than between mountains and plain, and certainly much exchange of culture 
took place, but nonetheless the special relationship with the mountains from whence 
the Elamites probably originated, made Khuzistan quite different from Mesopotamia. 
The interchange of products between the cool mountains and the hot plains was 
ancient in origin and continued throughout millennia. Many fruits and flowers were 
given to their western neighbors by the people of the plateau, and just the later Iranian 
words in Semitic languages alone such as Aramaic warda, 'rose,' testify to the 
exchange. Likewise, the time of the introduction of the date palm to the warm oases of 
southern Iran is unknown, but it is generally agreed that the earliest cultivation is 
found in southern Mesopotamia. Thus the ancient exchanges, though undocumented, 
can be assumed as a continuing process. So the geographical unity of Khuzistan with 
Mesopotamia, at least from the economic point of view, made less sense than the 
connection between the lowlands of Khuzistan and highlands to the east, and the 
occupation of both areas by Elamites, or at least Elamite culture, is understandable. It 
is true that the mountain ranges are very rugged and difficult of access, but this did 
not deter either merchants, or peoplejust seeking to escape the heat of summer on the 
plains. Surely ideas and artifacts were exchanged in this area from ancient times. 

The location of the mountain ranges, generally in a northwest— southeast direction, 
made descent from the plateau to the plains easier southward from Luristan than 
directly west to Babylonia. The inhabitants of Luristan were those mountain dwellers 
who had most continuous contact with Khuzistan. In the many valleys of Luristan 
one may postulate an early Iranicization of the original inhabitants who were 

For example, the word for 'stone' is bard both others. Many words and expressions are not 
in Luri and Laristan dialects, also 'large' gap and peculiar to these two groups of dialects, however, 
gapu respectively, 'egg' xa and xag, and many but they are not used in Persian. 



Demography 29 

probably a mixture of Elamites, Quti or Guti, Kassites and others. The predominantly 
pastoral economy of the Lurs today must have been similar over millennia, and a 
multitude of small settlements was the rule, rather than any large cities in the region 
we call Luristan. 

The Lurs years ago were usually divided into two groups, the large (buzurg) and 
the small (kuchek) Lurs, the former located roughly to the southeast and east of the 
Ab-e Diz and the latter to the northwest and west of the upper reaches of that river. 
The small Lurs were divided primarily into two geographical areas, Pish-e Kuh 
(before the mountain) and Pusht-e Kuh (behind the mountain), the range being the 
Kavar or Kabir mountains. The former region includes the chief city of 
Khurramabad, but today much of the population in the northern valleys, up to 
Kermanshah speak Lakki dialects which are south Kurdish dialects. The Pusht-e Kuh 
population lives in wilder surroundings and has been less studied than their northern 
cousins. The Luri dialects generally spoken here merge with southern Luri of the 
Lur-e buzurg. The main tribes are the Mamasani, Kohgiluye, Boyer Ahmad and 
Bakhtiyari, the last formerly divided into the Haft Lang and the Chahar Lang. 
Whether the present divisions of the Lurs reflects an ancient ethnic differentiation is 
unknown, but it is not improbable, for the rugged geography of Luristan alone 
would make it an excellent divided refuge area. 

The present linguistic picture does not help us with reconstructing the past. As 
noted, speakers of Lakki dialects in northern Luristan are probably relatively late 
intruders, as are the Kurds in their homeland to the north of the Kermanshah- 
Hamadan road. One may assume that the ancient population must have been as 
divided as the present, but there is some archaeological evidence that at least culturally 
the area did have a unity. Also the Iranicization of the ancient peoples of Luristan- 
Guti, Lullubi, Kassites and others seems to have been uniform, in that ancient 
differences do not appear in the modern dialectology or ethnography of Luristan. 8 In 
other words, Luristan as an entity, with a common culture and group of dialects, is an 
ethnographic as well as a geographic and historical reality. The origin of the word 
Lur is uncertain, but folk tradition claims it comes from the gypsies or Luris brought 
from India by the Sasanian king Bahram Gor, as reported in Firdosi's Shahname, to 
bring music to the common people of Iran. Usually these itinerant musicians are 
called Luli, but it is possible that the name is a transfer to the inhabitants of the 
mountains. In any case, the word is not an ancient designation for the land or its 
peoples. 

Whereas the Lurs and their dialects are closely related to the Persians of Fars 
province, and naturally belong to the southwestern branch of the Iranian peoples, the 
Kurds are more complicated and their relations are much in dispute. It is generally 
accepted that the Kurdish language in its two major divisions of Kurmanji (northern) 
and Sorani (southern) is related to eastern and central Iranian dialects, thus indicating 
an eastern origin in respect to the geographical position of Kurdish today. 9 The 

8 On Luristan and Luri dialects see the works of Windfuhr, "Isoglosses: A sketch on Persians, 

HamTd IzadpanSh, esp. his Farhange Luri (Tehran, Parthians and Medes," AI, Hommages et Opera 

1 343/1 965), and 'Ah ftasan, Gozaresh-e Govlshha- Minora Monumentum H. S. Nyberg, 2 (Leiden, 

ye Luri (Tehran, 1342/1964). 1975), esp. 470-71. 



9 



See the survey, with references of G. L. 



30 Chapter II 

Median dialects which were absorbed or supplanted by Kurdish dialects may be 
considered as represented today by Zaza and Gorani, both of which, however, were 
heavily Kurdicized. The process of Kurdicization of both the Gorani speakers (now 
living in the mountains north of the Kermanshah-Hamadan road) and of the Lakks 
(south of the road) must have slowly proceeded throughout the Islamic period. 1 ° The 
Zaza speakers in Anatolia also have a tongue different from Kurmanji and Sorani, 
even though one cannot thereby connect closely Zaza with Gorani. The native 
designation of the Zaza language as Dimli has been interpreted as coming from 
Dailami, a Caspian dialect spoken by the Dailamis, who expanded from the Elburz 
mountains to the upper Zagros mountains, in the eleventh century. 1 i Whether this is 
true or not, the great difficulty in separating various Iranian dialects and languages 
from each other in the general area of Kurdistan is obvious, since many strata and 
overlays have occurred because of migrations or forced settlement of groups from 
elsewhere. Just as in the Hindukush mountains of the east, so here too we find an 
ethnographic and linguistic museum of various peoples, again both refuge areas. 

In Kurdistan, as in Luristan, it is hardly possible to draw ancient conclusions from 
the present distribution of dialects or religious sects, except to note that geography 
also here plays an important role in determining where different peoples live and 
lived. The land dictates either agriculture in the valleys or pastoralism on the slopes of 
the mountains, and it is possible that the extensive spread of pastoralism is the result of 
the spread of Iranian-speaking tribesmen into those areas. Of course, we have r/o 
written or archaeological evidence for this supposition, but the Iranicization of the 
northern Zagros range may well have proceeded from occupation of pastoral 
meadows to the settled agricultural land. 12 Obviously the process of Iranicization 
took time, but the lack of any apparent ethnic or linguistic unity in ancient 
northwestern Iran, not to speak of a political empire, probably aided the spread of 
Iranian speech. Together with several presumed confederations of tribes, the Iranian 
dialects seem to have presented a greater unity among the invaders than was found 
among the aborigines. All indications from Assyrian sources, which tell of Assyrian 
forays upon the Iranian plateau from the beginning of the first millenium B.C., reveal a 
diverse linguistic, cultural and presumably ethnic, mosaic among the aboriginal 
population here. Also we may further presume that the Iranian dialects were spread 
over relatively large expanses of territory and were insofar mutually intelligible that 
one might conclude that Iranian easily became a kind of lingua franca not only for 
Iranian pastoralists, but also for the settled people on the plateau with whom the 
Iranians came in contact. As mentioned above, we may further infer that the Kurds 
are still in the process of expansion from the highlands into lowland villages of the 
plain of Urmia and elsewhere, whereas the Zaza and Goran are remnants of a settled 



10 G. B. Akopov, "Kurdy-gurany," Strany i Persische Forschungen (Berlin, 1930), 14-17, and 

narody blizhnego i srednego vosloka 7, Kurdove- in the same series his Mundarten der Zizi (Berlin, 

denie, Akad. Nauk Armyanskoi SSR (Erevan, 1932), 2-6. 
1975), 164-65. i^Cf. the discussion in Vilchevskii, op. cit. 

1 ' On the Goran and Zaza see K. Hadank, supra, Kurdy, 25-39. 
Mundarten der Guran of O. Mann's Kurdisch- 



Demography 31 

Iranian-speaking population which had settled in mountain villages before the 
Kurdish expansion. 13 

Azerbaijan, strictly speaking, comprises the valleys and plains between the 
mountains of Kurdistan in the west to the Elburz mountain range descending to the 
Caspian Sea in the east. Even today in the western highlands, where the towns of 
Sanandaj, Bijar, Sakkiz and Mahabad are located, the majority of the population is 
Kurdish-speaking, whereas just to the east, in the lower lands, between the towns of 
Takab and Miyandoab the population is Turkish-speaking. Does this present division 
correspond to an ancient one? On the basis of scanty source material it is not possible 
to say whether present divisions could reflect earlier ethnic or linguistic 
differentiation; the differentiation, generally speaking, would be between predomi- 
nant pastoralists in the mountains and agriculturists in the lowlands. In the Assyrian 
annals, the area comprising the present province of West Azerbaijan was divided 
between the Urartians in the north and Manneans to the south of Lake Urmia, while 
the area of Urmia itself was probably a place of mixture. No ancient information is 
available, however, on the mountain-lowland division, for possibly relevant names 
in the ancient geographers Strabo and Ptolemy cannot be localized. The lake was 
perhaps the best known landmark of Azerbaijan, called the lower sea of Nairi by the 
Assyrians and later the lake of the Mantiane (Manneans) or the "blue" lake (Kabodan) 
and also Chaechist by the Iranians. 14 

Many scholars have tried to identify the place names and tribal names in Assyrian 
sources but most of the localizations are only guesswork. The general reconstruction 
of the historical geography of Azerbaijan by I. M. Dyakonov, Istoriya Midii 
(Moscow, 1956), 87-93, is accepted by most scholars with modifications made by L. 
Levine, "Geographical Studies in the Neo-Assyrian Zagros," Iran, 1 1 (1973), 1—28 and 
12 (1974), 99-122. According to Dyakonov, the area called Sangibutu by the 
Assyrians was the contemporary fertile Khoi-Marand area, which was conquered by 
the Urartians in the ninth century B.C. To the east of this region was the valley of the 
Kara-su, a southern tributary of the Araxes River which was later occupied by a 
people called by Classical authors the Cadusii. The identification of other localities or 
tribes in the Assyrian sources is fraught with many difficulties and reveals the 
complexity of the historical geography of the land. 

While there is no evidence of ethnic or linguistic unity among the pre-Iranian 
population of Azerbaijan, we may guess that some of them had relations with small 
groups of Caucasian peoples of today who in ancient times extended farther to the 
south. Thus, the ancient Albanians who lived in present day Soviet Azerbaijan, may 
have been the descendants of what might be called linguistically Caspian peoples 
living on the western and southern shores of the sea, whereas to the west was the 



1 3 Cf. Hadank, loc. cil. [n. 1 1 ]. The Mukri Kurds Kurdish dialects are different from those to the 

who today occupy the area of Mahabad and south in Sanandaj and Kermanshah, and this 

Sakkiz, may have migrated from Iraq in the 17th division may possibly represent an ancient division 

century according to O. Mann, Die Mundarl der of peoples. 
Mukri-Kurden, 1 (Berlin, 1906), xviii. The Mukri l4 Strabo XI, 523 and XI, 529. 



^2 Chapter II 

Hurrian-Urartian and possibly Mannean linguistic group. 15 We do not know how 
far to the north Elamite dialects or related languages extended, nor indeed how many 
other tribes or peoples existed, speaking tongues different from the two main groups 
in western Iran. In any case, at the time of the expansion of the Iranians we may 
presume that western Iran was dominated by the Elamites in the south and Urartian- 
Hurrians in the north. To the east, however, we have no sources other than 
archaeology. 

Archaeology has revealed highly developed cultures in the south Caspian Sea 
regions, as early as the second millennium B.C., at sites such as Marlik and Kaluraz in 
Gilan. 16 Archaeological surveys have indicated that the Gilan plain probably was not 
settled before the Parthian period, and that all ancient sites are located in the foothills 
of the Elburz mountains, none of them of great size indicating a sparse population. 17 
Since no early second millennium or earlier pottery or other ancient objects have 
been found in Gilan or western Mazandaran, we may conclude that the Iranians were 
the first people to settle in the south Caspian plains in any numbers. In the Talysh 
provinces, however, especially in the Soviet Union, ancient graves have been found 
dating from the second millennium B.C. Presumably the descendants of these people 
were the Cadusii, about whom we know little more than their name. Yet Strabo (XI, 
508) associates them with the Anariaki or "non-Aryans," and we may suppose that the 
Cadusii were non-Iranians and related to the Albanians or people of Arran, as it was 
called in the Islamic period, who were then Iranicized. 

The plateau where today are located the cities of Tehran, Hamadan and Isfahan, 
was the center of Iranian settlement, but who were the aborigines here whom the 
Iranians conquered or absorbed ? We know from archaeological excavations at sites 
such as Tepes Giyan, Hissar and Siyalk that there existed cultures before the coming of 
the Iranians. The Iranicization of the central plateau seems to have been so complete 
that no conclusions may be drawn about the ethnic or linguistic nature of the 
autochthonous population. Pottery and other remains do not help in this regard, but 
we may suppose that the population on the plateau was sparse, for as mentioned 
previously, techniques of irrigation such as the underground water channels are not 
attested before the first millennium B.C., which enabled agriculturists to extend 
cultivation into the plains and deserts, thus opening virgin tracts of land to 
settlement. 18 On the other hand, some oases have lost their water supplies and have 
become desert, as at ancient sites such as the graves of Shahdad in the Lut desert to the 
northeast of Kerman where pre-Islamic remains have been found. 19 Who were these 
pre-Iranian people? We may suppose that the inhabitants of the areas to the south and 



The close relationship of Urartian and l7 Ibid., plus A. Hakemi, "Excavations in 

Hurrian has been established, but with no written Kaluraz, Gilan," Bulletin of the Asia Institute of 

remains of the Manneans, the nature of their Pahlaui University, 3 (Shiraz, 1973), 1-3. 

language is conjectural. See I. M. Dyakonov, ' 8 See ch. 1 , n. 37. 

Yazyki Drevnei Prednei Azii (Moscow, 1967), 19 Cf. M. A. Kaboli, "Shahdad in the Third 

113-20. Millennium B.C." (in Persian), Proceedings of the 

Cf. Hrouda, op. cit., [p. 26] Vorderasien, 248. Hnd Annual Symposium on Archaeological Research 

Although the graves may extend into the first in Iran (Tehran, 1974), 1-9, and reports by A. 

millennium and the period of Iranian invaders, the Hakemi at the Vllth International Congress of 

art forms are autochthonous. Iranian Art and Archaeology at Munich in 1976. 



Demography 33 

east of the Lut desert in Kerman, Seistan and Baluchistan may have been related to 
each other and possibly to others both to the north in present Turkmenistan or to the 
southeast in Pakistan. 20 Certainly trade existed from a very early period, even in 
luxury items such as lapis lazuli from Badakhshan in Afghanistan, for it has been 
found in early graves in Mesopotamia. 21 The urban settlements, however, must have 
been small and few, first merely as suppliers both to peasants and to pastoralists, and 
then developing into centers where craftsmen produced their wares as specialists. It is 
important to remember in the history of Iran that although agriculturalists and 
pastoralists represented two different and sometimes inimical ways of life, a third 
factor, urban dwellers, were important in their own right. Invaders or nomads 
usually tried to control the towns of a land, where they interacted with townspeople. 
On the whole they despised the villagers and did not intermarry with them. The 
towns were stages where events were enacted and records were compiled, so history is 
their story. Unfortunately we know very little about large settlements on the Iranian 
plateau in ancient times, and even less about the composition of the population of 
those centers. There were many racial types as seen from wide, long and other types of 
skulls from burials, and one may presume that mixtures of populations had been 
occurring since palaeolithic times. One must constantly re-emphasize the vital 
importance of water to any life, settled or nomadic, especially in eastern and 
southeastern Iran, which were barren wastes, and archaeology reveals constant 
attempts at building dams or cisterns to preserve every drop of the precious fluid. 22 
Under difficult physical conditions the transient nature of settlements in eastern Iran 
and Baluchistan is not surprising, since settlement patterns followed the availability of 
water and the rise or fall of the water table. One should not expect many large towns 
in the east, and it would seem that pastoral and nomadic life was dominant here, 
except in the province of Seistan where the Helmand River provided water for 
extensive agriculture. Today the population is a mixture of Persians, Pashtuns, Baluch 
and even Brahui, and we may presume that the region was also mixed in antiquity. 
The Brahuis, who speak a Dra vidian language, were either offshoots of a wandering 
Dravidian tribe from the Deccan, or they represent an ancient branch of Dravidians, 
who may have come from the sub-continent at the end of the fourth millennium B.C. 
The latter theory is generally accepted today, because of the archaic nature of Brahui 
among the Dravidian languages. 23 If this is the case, then the proto-Brahuis may have 
been the dominant people in Baluchistan and even Sind when the Aryans expanded. 
Whether they were the inhabitants of Mohenjo-Daro and carriers of the Indus 
civilization, probably destroyed by the Vedic invaders of India, is uncertain. 
Although the Iranian-speaking Baluchis came from the west in Islamic times, before 
their arrival other Iranians had mixed with the Brahuis, as can be seen from archaic 

20 M. Tosi, "Shahr-i Sokhta: a charge and a Phases of its Trade," Iraq, 30 (London, 1968), 29- 
chance for urban archaeology," in F. Bagherzadeh 30. 

ed., Proceedings of the 1st Annual Symposium of 22 Cf. J. T. Maruchek, "A Survey of Seasonal 

Archaeological Research in Iran (Tehran, 1973), 1- Occupation Sites in Northern Baluchestan," in 

5; also his "A Topographical and Stratigraphical Baqerzadeh ed., Proceedings of the IV th Annual 

Periplus of Shahr-e sukhte, "Proceedings of the IVth Symposium on Archaeology (Tehran, 1976), 272. 
Annual Symposium (Tehran, 1976), 43. 23 M. S. Andronov, Yazyk Braui (Moscow, 

21 G. Herrmann, "Lapis Lazuli: The Early 1971), 12-13, with a bibliography. 



34 Chapter II 

Iranian borrowings in the Brahui language. 24 Also the extent of Brahui settlements in 
early times cannot be determined, but one reason they have not been absorbed by the 
Baluchis was the primarily nomadic-pastoral way of life of the latter as contrasted 
with the settled Brahui of Kalat and elsewhere. If much of the Iranian national epic, as 
preserved in Firdosi's Shahname, can be traced to Seistan, then the identification of the 
Brahuis as the Turanians of the Epic might be considered. It would seem, however, 
that other peoples also lived in this desolate part of Iran, and today's dialects may 
indicate also an earlier linguistic division. In the highlands of Bashakird (or Bashkard) 
to the northeast of Minab live primitive agriculturists speaking Iranian dialects, while 
to the east in the lowlands of Bampur and beyond live Baluchis, most of whom are 
nomads or semi-nomads. Along the coast are fishermen, a medley of races including 
Negritoes, and negroids, the former from Malaysia to the east and the latter from 
Africa. One cannot tell whether they are relatively recent arrivals brought as slaves, 
or whether they (especially the Negritoes) may represent an ancient strain in the 
population. 25 Classical authors tell offish-eaters (Ichthyophagi) and other folk living 
along this coast, but their ethnic or linguistic identity, of course, cannot be 
determined. 

In early Islamic times the people of the interior of Bashakird and Makran were 
called Qufs and Balus by Arabic authors, the former presumably mountaineers and 
the latter pastoralists. 26 Again the differentiation of peoples by their mode of 
livelihood is striking, and a possible ethnic differentiation in antiquity is possible. 
Today, of course, all of the population speak Iranian languages or dialects, and we may 
assume that the Iranicization of the earlier inhabitants was not a prolonged process. 27 
Along the southern coast only the harbors at Bandar 'Abbas and Tiz-Chahbahar 
offered access to an interior which could export products, in the former case Sirjan 
and the province of Kerman, and in the latter Bampur and the Saravan region of 
Baluchistan. 

Finally, a few words about the population of the central deserts of Iran may aid us 
in placing these people in the history of Iran. Inasmuch as the domestication of the 
camel long preceded the arrival of the Iranians on the plateau, we may presume that 
the oases of the central deserts of Iran had been long occupied, at least as way stations 
on trade routes from east to west, and the archaeological finds at Shahdad in Kerman, 
mentioned above, would tend to substantiate this hypothesis. Unfortunately no 
excavations or even archaeological surveys have been carried out in any of the oases, 
but Ptolemy (VI, 5, 1) mentions a district of Parthia called Tabikene, which one 
supposes is the desert area of Tabbas. Also Stephan of Byzantium, in his Ethnika 
mentions the Tabenoi as a people living in the desert of Kerman, so we may assume 
that the oases were inhabited. The numbers of people living in such isolated places 
even today is small, and in antiquity it cannot have been more, so we may suppose 
that the early inhabitants, like today, made their living on date palms and as watering 
places for caravans. It should be noted that ancient copper mines have been found in 

24 A. V. Rossi, "Iranian Elements in Brahui," Bosworth, The Kufichis or Qufs in Persian 
AION, 31 (1971), 401. History," Iran, 14 (1976), 9-17. 

25 Cf. Pikulin, supra, Beludzhi, 31. 21 Cf. the various articles of I. Gershevitch on 

26 B. Spooner, "Koch u Baloch and Ichthyo- the dialects of Bashakird (or Bashkardia) listed in 
phagi," Iran, 2 (London, 1964), 53-59, and E. Bosworth, op. cit. [n. 26], n. 20. 



Demography 35 

the central desert near Anarak, and one may further suggest that this area was one of 
the sources for the metal in antiquity. 28 The population probably did not differ from 
the settled people in the Qom and Kashan areas to the west. Tepe Siyalk is not far from 
the copper mines. 

In summary, the population of Iran in antiquity seems to have followed much the 
same pattern as in the recent past. Where pastoralists roamed, conditions of geography 
and hence life have little altered, and where villages barely maintained an existence 
with scarce water supplies, conditions have remained much the same without great 
changes. The mountains were refuge areas with little contact from one valley to 
another, and peasants were oppressed either from the towns or from the pastoralists 
(or nomads), and sometimes from both. There were three centers which could be 
unified and provide bases of power or expansion, Fars province in the south, 
Azerbaijan in the north, or the central plateau area of Hamadan-Tehran-Isfahan. 
Other parts of Iran were either dependent on one or more of these areas, or lived in 
relative isolation. But from very early times trade and commerce were important 
across Iran, as well as inside, and the age-old invasion route into Iran from Central 
Asia was used by nomads seeking better pasture lands. Since the pattern of migration 
seems to have been invariably from east to west, we should turn to the east and north 
to examine the inhabitants of those areas. 

AFGHANISTAN 

As an extension of the plateau to the east, conditions of life in Afghanistan are similar 
to those in western Iran, although in the east water is scarcer and the mountains are 
higher than in the west. The inhabitants of Herat, stretching to ancient Bactra (Balkh) 
on the east, were an extension of the population of the present-day Iranian province of 
Khurasan, and from the time of the Achaemenid Empire these 'upper provinces' had a 
unity, probably ethnic as well as cultural. As usual, it is in the mountains that one finds 
small groups of people the very existence of whom can perhaps tell us about the 
settlement or movement of peoples in the east. For a pre-Indo-Aryan population, we 
should consider the Burushaski people of Hunza, Pakistan, for they, like the Brahuis 
to the south, may be yestiges of an earlier population which was much more 
widespread than today. Unlike Mesopotamia, the vast subcontinent of India has very 
few ancient written records which can be used for historical purposes. Consequently 
there is no information about the past of the Burushaski speakers whose language has 
no relation to its neighbors of the Iranian, Indian or Tibeto-Burman families. 29 

The pre-Aryan past of the east is much darker than in the west, and just ns the proto- 
Brahui hypothesis is guesswork for the southern part, so that of proto-Burushaski 
speakers is hypothetical for the northern part of the Indo-Iranian borderlands. It must 
be remembered that nomadism and semi-nomadism are much more widespread in 

28 Wertime, op. cit. supra, Metallungy, 1257-67. 29 On the Burushaski language see D. L. R. 

Other mines, dating from the Bronze Age, have Lorimer, The Burushaski Language, 3 vols. (Oslo, 

been found near the city of Qom; cf. H. Holzer, 1935), and H. Berger, Das Yasin-Burushaski 

Ancient Copper Mines in the Veshnoveh Area, (Werchikwar), (Wiesbaden, 1974). 
Kuhestan-Qom," Archaeologica Austriaca, 59 
(Vienna, 1971), 10-11. 



36 Chapter II 

eastern than in western Iran, and the settled folk in the east have been more subject to 
domination by tribes than in the west. Whether proto-Burushaski speakers were 
nomads who were pushed into the mountains by invading Indo-Aryans, or whether 
they were always settled mountaineers cannot be determined. In the east, unlike the 
west, traces of the first wave of Indo-Aryans, comparable to those in the Mitanni 
confederation in the west, still survive. These are the present speakers of Dardic and 
Kafir (Nuristani) languages. 

The Dardic peoples today live in the mountainous northeastern part of Afghanistan 
and the northwestern part of Pakistan and India. The languages occupy a position 
between Indian and Iranian, and in spite of disagreements about the origin of these 
languages, all investigators agree that the Dardic languages can be considered a 
separate group of Indo-European tongues with features particular to that group. 30 In 
the old religion and folklore of the Dardic peoples one can find many features which 
support the thesis that these people were much more widespread in the past than 
today, and that they represent the remains of the earliest wave of Indo-European 
speakers who entered the plains of India. 31 If we assume that linguistically this 
suggestion makes sense, then these pre-Vedic Aryans may have carried a distinctive 
black pottery with them as proposed by R. Ghirshman. 32 We may assume then that 
these early Indo-Aryans were pushed into the mountains by the Vedic Aryans 
invading India from the northwest about the twelfth century B.C.; but did the Dardic 
speakers constrain the aborigines to retreat even further, so that today we only have 
the Burusho, the Burushaski speakers, as their remnants? If this is true, there seems to 
have been relatively less absorption of the previous inhabitants by newcomers than in 
the west. For we must remember that the final movement into the Indo-Iranian 
borderlands, in Islamic times, is the expansion of the Iranian Pathans or Pashtuns from 
their homeland in the mountains of Afghanistan onto the plains of the northwest 
subcontinent. 

The Pathans are organized tribally and rule present Afghanistan. They are the 
pastoralists par excellence of the Indo-Iranian borderlands, practicing an annual 
migration to the highlands of Afghanistan in the summer and to the lowlands in the 
winter. Herodotus (IV, 44 and VII, 67, 85) mentions the land of Paktyia and the 
Paktyes living near Kabul, which may be the ancestor of the name Pakhtun or 
Pashtun, used by the nomads. The Pathans expanded from their homeland in the 
Sulaiman mountain range onto the plains only in the aftermath of the Mongol 
invasions in the twelfth and thirteenth centuries of our era, and it would seem that the 
Pashtu language is today still expanding in Pakistan. Legends linking the Pashtuns to 
the Hephthalites or to the Khalaj Turks may reflect a northern origin, although 
nothing else can be said about their origins. 33 Pashtu, probably, has absorbed other 

30 On the Dardic languages see D. I Edelman, also his Die Sprache von Sau in Ostafghanislan 
Dardskie Yazyki (Moscow, 1965) with an exten- (Munich, 1967), ll,andK.Jettmar,Di'ei?f%'o«e« 
sive bibliography, and on the inhabitants P Snoy, des Hindukusch (Stuttgart, 1975). 

Die Kafiren (Frankfurt, 1962), 10, and G. G. 3: R. Ghirshman, L'han el la Migration des Indo- 

Morgenstierne, "Die Stellung der Kafirsprachen," aryans el Iraniens (Leiden, 1977), 68-69. 
lrano-Dardica (Wiesbaden, 1973), 327-43 " G. Morgenstierne, in Iranislik, Handbuch der 

31 G. Buddruss, "Zur Mythologie der Prasun Orientalistik, (Leiden, 1958) 169, also his Parachi 
Kafiren," Paideuma Festschrift ftir H. Lommel, and Ormuri [n. 35, infra], 8. 

(Wiesbaden, 1960), 207, with Vedic parallels. Cf. 



Demography 37 

Iranian speakers in Afghanistan as well as Indie and Dardic speakers in Pakistan, and 
two survivors of older Iranian languages, now almost wholly engulfed by Pashtu, are 
Parachi and Ormuri, the former of which is still spoken in a few isolated villages of 
the Hindukush mountains north of Charikar, while the latter is still spoken by a few 
people in the Logar Valley south of Kabul and in a village in Wazirstan, Pakistan. It is 
fascinating that both of these languages show many features in common with Persian, 
and they have been designated as southeast Iranian relict languages. 34 Unfortunately 
it is not possible to determine when the ancestors of the Parachi and Ormuri speakers 
arrived in the mountains of Afghanistan, and exactly what relationship they had to 
the Pashtuns. If the relations between their tongues and New Persian could be 
clarified, the proto-Parachi and Ormuri speakers might have been an intrusion from 
the west in historic times. Or they might have been the original Iranian invaders into 
a Dardic-speaking area in very early times. A few concurrences in vocabulary 
between these languages and Brahui, Baluchi and Pashtu indicate little more than 
geographic proximity and complex borrowings which cannot be tied to historical 
events or even to the movements of peoples. 35 

When we turn to the Pamir languages, their isolated geographic location, north of 
the Hindukush, has caused much inter-borrowing, making reconstructions of 
etymologies and family relationships difficult. While some scholars have sought to 
derive one or more of the Pamir languages from the Middle Iranian Khotanese Saka, 
it seems better, from what little we know, to speak of an unknown proto-Pamir 
language as the ancestor of the present Pamir languages. 36 At least some 
differentiation has been made among the Pamir languages, and the Shughni group of 
languages (Roshani, Bartangi, Sarykoli, and Yazghulami) seems to constitute one 
family, while Munjani, Wakhi, Ishkashmi, and other little known languages, 
constitute other families or relationships. 37 When such words for 'sun,' as Ishkashmi 
remuz and Sanglechi ormozd (from ancient Ahura Mazda), and Munjani mlra from 
Mithra, occur in the Pamir languages, it readily will be seen that the folklore and 
beliefs of the Pamir peoples, as well as their languages, have interesting implications 
for reconstructing the past of the east-Iranian world. 

Whereas the Pamir languages cannot be considered as descendants of any known 
ancient Iranian language, the language of the Yaghnobis, today less than two 
thousand, living in an isolated mountain valley a hundred kilometers to the north of 
Dushanbe, Tajikistan, is descended from a Sogdian dialect. The Yaghnobis are the last 
remnants of a once populous folk who inhabited the valley of the Zarafshan River, 
where the cities of Bukhara and Samarqand are located. Until fragments of 
manuscripts from Chinese Turkestan written in the Sogdian language were 
deciphered and identified in the beginning of the twentieth century, it was not 
known what kind of a language Yaghnobi was. 38 But Yaghnobi and the Yaghnobis 

4 Morgenstierne, Iranistik, hoc. cit. on linguistic Shughni Group (Wiesbaden, 1974), with an 

grounds, suggests that Pashtu may have originally extensive bibliography, 

come from the north. 37 Cf. T. N. Pakhahna, Pamirskie Yazyki 

Morgenstierne. Indo-lranian Frontier I.an- (Moscow, 1969), 9-10. 

guages, 1, Parachi and Ormuri (Oslo, 1929), 13, 38 See A. Khromov, Yagnobskii Yazyk (Mos- 

3 "- cow, 1972), with bibliography. 
Morgenstierne, Etymological Vocabulary of the 



38 Chapter II 

are important not only from the point of view of linguistics but also for anthropology 
and folklore. 39 The Sogdian language first gave way to Persian, and then in many 
parts of Russian Turkestan Persian/Tajiki was supplanted by Turkic languages. In 
reconstructing the life of the common folk of Central Asia before the Arab conquests, 
the Yaghnobis can be helpful, especially in comparing words used today with 
Sogdian. It must be emphasized that Islam and the new Persian/Tajiki language have 
both greatly influenced the Pamir peoples and the Yaghnobis such that it is very 
difficult to unpeel the accumulation of centuries to try to recover the past. 

The Yaghnobis, of course, are not the only Iranian speakers in Central Asia, for the 
vast majority are now Tajiki/Persian speakers, who mostly live in Tajikistan, with its 
capital Dushanbe. In antiquity the people of Tajikistan were called Bactrians, and they 
extended also to the south of the Amu Darya into Afghanistan where their main city 
was located, the present-day ruins of Balkh. So the ethnographic, if not also linguistic, 
unity of the Bactrians living south of the Hissar mountain range, seems apparent from 
earliest times. To the north of the Hissar range lived the Sogdians, mostly in what is 
now Uzbekistan. Just before the Arab conquests, we infer that Buddhism was the 
dominant religion and culture of the Bactrians, while a local form of Mazdaism or 
Zoroastrianism seems to have been prevailing among the Sogdians. But the Sogdians, 
an active trading folk, had trading colonies as far to theeast as China and Mongolia, 
and many religions, including Christianity, Manichaeism and Buddhism, were 
popular with the Sogdians. The Sogdians also were very mixed ethnically because of 
their strategic location, as were their cousins in Khwarazm, the land on the lower 
course of the Amu Darya south of the Aral Sea, who spoke a language different from 
Sogdian. Although the Tajiki dialects of today, unlike Yaghnobi, are descended 
from New Persian, nonetheless many Sogdian survivals may be found in them, 
distinguishing Tajiki from Persian spoken in western Iran. Many studies of Tajik 
folklore and anthropology, as well as dialectology, help us further to reconstruct the 
past of that part of the world, especially old forms of family relationship and social 
life. 40 

The Turkic speakers, Uzbeks, Turkmens and others, are of lesser aid in 
reconstructing the past, since they have added a layer of Turkic traditions >and 
practices to that of the Tajiks, and are thus further removed from the culture of the 
ancient inhabitants of the land. Only in Turkmenistan are, and were, nomadism and 
pastoralism dominant over the settled life, and so in antiquity the steppes of Dahistan, 
to the east of the Caspian Sea, also supported a nomadic population. According to 
Classical authors in antiquity here lived the Massagetae, an Iranian people who raided 
settled lands to the south as did the Turkmens in recent times. Again, here as in Iran, 
climatic conditions have not changed much since antiquity. 

Paleoanthropology tells us that the population of Central Asia in antiquity was 
much the same as the south of the Amu Darya with mixtures of long-heads and 
mesocephalics; while possible connections with the south may be found in certain 

39 M. S. Andreev, Malerialy po etnograjii Yag- works on the folklore of the peoples of the Pamirs 
noba (Dushanbe, 1 970). and much research has been done on various Tajiki 

40 The Institute for the study of the Pamirs in dialects and folklores. 
Dushanbe has published a number of valuable 



Demography 39 

cranial types similar to the Dra vidians of India. 41 Man, of course, lived in very early 
times in Central Asia, for Neanderthal remains have been found there, including the 
skeleton of a boy of eight or nine years old. 42 It is unnecessary to go back into 
prehistory, however, since neither cranial types nor various prehistoric artifacts aid in 
the reconstruction of later history. Suffice it to say that archaeology is yearly 
revealing very early settlements in Central Asia before the expansion of the Indo- 
European-speaking peoples from south Russia, and the pattern of life found there is 
much the same as in western Iran, with the settled population more numerous than the 
nomad-pastoralists, except in Turkmenistan. To the north of the present Tashkent 
area, however, the steppes were dominated by nomads. 

The Kara Tau and Kirghiz (formerly Alexandrovskii) ranges were the dividing 
line between the oases towns of Russian Turkestan (today the Soviet republics of 
Turkmenistan, Tajikistan and Uzbekistan) and the vast steppe lands of present 
Kazakhastan and Kirghizia. The population of the steppes when we first learn about 
them, was naturally nomadic except in the southern forested valleys of present Alma 
Ata and Frunze. To the far north were the forests and tundra of Siberia, the domain of 
hunters and fishermen. Mobility was the key to life on the steppes, and the almost 
paramilitary organization of the tribes who roamed here, made them formidable 
opponents of any settled folk. Slowly archaeology is also uncovering the material 
civilization of nomads whose culture was more complex than hitherto supposed. We 
should pause to consider briefly the origins of pastoralism. 

When we think of pastoralists as horse-riding nomads, it refers to historic times, for 
even the early Indo-European migrants were chariot-riding invaders into India, as also 
the Myceneans into Greece. The use of horses primarily as riding animals rather than 
pullers of wagons cannot date much before the middle of the second millenium B.C., 
and it took a long time before riding displaced the use of horses as draught animals. 
On the other hand, true nomads could hardly exist without the use of horses 
primarily for riding, and this suggests that nomadism is a relatively late and 
sophisticated development of a pastoralism based on the use of wagons to follow the 
herds. 43 Wherever and whenever the domestication of the horse took place, possibly 
in the Near East, it was nonetheless the Indo-European-speaking peoples who later 
utilized the horse more than any others, and it surely contributed to the spread of 
those peoples. The Indo-Europeans in the course of their wanderings, with chariots 
and wagons, may have learned that riding was more efficient than chariots, especially 
in combat. 44 In any case, the first horse-riding nomads in history were the Iranian 



41 T. A. Trafimova, Drevnee Naselenie Khor- Central and Eastern Europe," American Schools of 
ezma po dannym paleoantropologii (Moscow, 1 959), Prehistoric Research Bulletin, 25, Peabody Museum 
1- (Cambridge, Mass., 1968), 5-7. 

42 For a bibliography of works on the Neander- 44 W. Brandenstein, "Das Pferd - eine Haupt- 
thal site of Teshik Tash in Uzbekistan by A. P. frage der indogermanischen Altertumskunde," 
Okladnikov, see Istoriya Vzbekskoi SSR, 1 (Tash- Mitteilungen der anthropologischen Cesellschafl in 
kent, 1955), 12-15 and 532 Wien, 92 (1962), 30-33; and J. F. Downs, "The 

43 Cf. F. Hancar, Das Pferd in prahistorischer und Origin and Spread of Riding in the Near East and 
Jriihhistorischer Zeit, Wiener Beitrage zur Kul- Central Asia," American Anthropologist, 63 (1961), 
turgeschichte und Linguistik, 1 1 , (Vienna, 1 956) ; 1 1 93-1 203. 

and S. BokOnyi, "Data on Iron Age Horses of 



40 Chapter II 

Scythians, or Sakas as the Persians called them, and they are found already into the first 
millennium B.C. The Scythians extended eastward to the forest lands of the Altai 
mountains and Mongolia, and it seems that the Turks and Mongols, who were 
originally forest dwellers, learned horse-riding from the Iranians. 45 The Scythians 
were carriers of a dynamic 'animal style' in their metalwork and on textiles, such as 
those preserved in the tombs of Pazyryk in Siberia. 46 In Siberia and the Altai 
mountains the Iranians mixed with Mongoloid peoples from the east, but the steppes 
were occupied by Iranian nomads throughout the first millennium B.C. Karl Jettmar 
has proposed, however, that horse-riding became dominant in Turkmenistan about 
1200 B.C. and afterwards was carried first to India and then northward to the steppes 
where previously agriculturalists had been dominant. 47 He further postulates a 
change from settled life to a half-nomadic pastoral existence and then suddenly to full 
nomadism, with large flocks and groups of warrior nomads. A comparison with the 
change in life-style of the American Indians after the introduction of the horse to 
North America, from a settled existence to buffalo-hunting nomads in the full light of 
history, can be instructive. 48 It would seem that the movement of peoples beginning 
about 1000 B.C. was the result, in large measure, of a great expansion in the use of the 
horse, first pulling carts and chariots and then as cavalry. We may conclude that from 
the period of earliest records in Iran, we find Iranian nomads in the steppes, and they 
were much later succeeded by Turks and Mongols. 

At the present time one small group of Iranian speakers remains from the once 
widespread Iranian world of the steppes. These are the Ossetes who today live in the 
Caucasus north of Tiflis, with their center at Ordzhonikidze (formerly Vladikavkaz). 
The Ossetes whose language has two dialects, a more archaic Digor, and the majority 
speaking Iron, are a classic refugee people. Although greatly influenced by their 
Caucasian neighbors, in customs, costumes and even language, it is generally agreed 
that they are descendants of the Alans, one of the Sarmatian peoples who ruled the 
south Russian steppes before the expansion of the Turks. 49 Because the Ossetes are the 
only living representatives of the northern Iranians, their language, and especially 
their folklore, has been studied to find ancient Iranian, or even Indo-European, 
features which survived in the refuge area of the north Caucasus. 50 The material 
culture of the Ossetes, however, is overwhelmingly Caucasian, since the Iranians 
settled down here so many centuries ago. 51 Although these northern Iranians were 
historically very important for the history of Eurasia, their impact on the Iranian 
plateau, the core area for this book, was only sporadic. Their history is only tangential 
to the history of the south, and we must omit any detailed discussion of them. 



45 See O. Lattimore, "The Geographical Factor 48 Ibid, 217. 

in Mongol History," Geographical Journal, 91 49 On the origin of the Ossetes cf. the results of a 

(1938), 8. conference on Proiskhozhdenie Osetinskogo Naroda 

46 On the finds of Pazyryk, see S. I. Rudenko, (Ordzhonikidze, 1967), where all points of view 
Frozen Tombs of Siberia (London, 1970), and S. V. about the origins of the Ossetes were aired. 
Kiselev, Drevnyaya Utoriya Yuzhnoi Sibiri (Mos- 50 See, e.g., G. Dumezil, Legendes sur les Nartes 
cow, 1951). (Paris, 1930), 9; also his Le livre des heros (Paris, 

47 Die fruhen Steppenv&lker (Baden-Baden, 1965), 11-12. 

1964), 215. si 5^ B A Kaloev, Osetiny (Moscow, 1971). 



Demography 4\ 

CONCLUSIONS 

We may summarize recent work on the ethnology and anthropology of the Iranian 
areas briefly. In the realm of physical anthropology and the study of crania, the 
presence of Negrito elements in southern Iran and Bachichistan, as well as Dravidian 
types in Central Asia in ancient times leads to the supposition that the pre-Aryan 
inhabitants of the Iranian plateau did in fact leave clues to their identity, but work in 
this domain has only begun. For example, no synthesis has been made, as far as I know, 
of the copious skeletal material from archaeological excavations, to portray the 
physical characteristics of the population in any area over a time period. Whether we 
can have any meaningful analyses of racial types in the history of the Iranian world 
remains to be seen, but at least overall syntheses are first needed. Studies on the 
dynamics of pastoral-settled occupation of land, and on types of coexistence have 
clarified the relationships of peoples with different ways of life. 52 As part of this 
question, investigations on the domestication of the horse have been mentioned as an 
important development in the early history of the Iranian world. The domestication 
and use of the camel was presumably less important here than it was in Arabia and 
North Africa, but the camel, like the horse and donkey, enabled man to go longer 
distances and to cross hitherto impassable deserts. Whether the camel was the decisive 
factor in causing the inhabitants of the Iranian plateau to abandon wheeled vehicles, as 
a recent study claims, is perhaps exaggerated. 53 It would seem rather, that the wheel 
had lost its importance long before the Sasanian era, as suggested in the study, and that 
the horse and donkey participated with the camel in relegating the wheel to 
inefficiency over the mountains and deserts of the Iranian world at least as early as the 
Achaemenids. 

One must always ask whether any contemporary model of ways of life can be 
projected profitably into an explanation of the past, ignoring changes in religion, 
culture, etc. For the vast majority of people throughout history the needs of survival, 
of economics, were far more important than intellectual activities, and continuity in 
nature, as well as way of life among the Iranians, suggests that contemporary studies of 
settlement and pastoral patterns do have great relevance to understanding the past; 
but in the part of the world we are discussing, the reverse is also true. Unfortunately 
we know very little about the instabilities of societies in the past which arose from 
failures to maintain subsistence levels. How many years of drought in a given area 
caused the inhabitants to move, or to change their way of life? It would seem that 
districts which could be characterized as frontier areas, on the margins of richer 
agricultural lands, were the prime centers of instability and were the scenes of 
transfers between settled and pastoral folk. Central Asia has been called the arena 
where tribal states ruled over vast distances, but only for relatively short periods of 
time, and this pattern overlaps as well into Iran and Afghanistan. On the whole, 



Cf. the works by F. Barth, such as his Indus 53 R. Bulliet, The Camel and the Wheel (Cam- 

aid Swat Kohistan (Oslo, 1956), his Ethnic Groups bridge, Mass., 1975), 175. 
<">d Boundaries (London, 1969), and W. Irons and 
N - Dyson-Hudson, eds., Perspectives on Nomadism 
(Leiden, 1972). 



42 Chapter 11 

people who lived in fertile oases, or valleys copiously watered by rivers, were stable 
and did not seek greener pastures elsewhere. Although history in this part of the 
world has been concerned with empires and kingdoms, many significant events, as for 
example the migration of the Baluchis, are tied to such matters as drought or plagues 
of locusts. Therefore, it behooves the historian to consider the contributions of the 
anthropologists and folklorists to our understanding of the past. 

In recent years the ancient and medieval city has caught the imagination of many 
scholars with the result that a plethora of studies and conferences devoted to the city 
has appeared. The oasis character of the Iranian world insured an early development 
of town life, and Soviet scholars have contributed much to our understanding of the 
Central Asian city, as well as to settlement patterns on the Iranian plateau. 54 The 
archaeological work in Central Asia, and in eastern Iran, at sites such as Shahr-i Sukhte 
in Seistan and Tepe Yahya in Kerman, has revealed ancient and extensive trade 
connections hitherto unknown. ss Eastern Iran and Central Asia are now revealed as 
separate cultural areas and not provincial extensions of western Iran. Furthermore, the 
oases of the plateau and Central Asia can make a claim to having been early centers of 
civilization, parallel to the river valleys of the Indus and Tigris-Euphrates, where the 
earliest written records have appeared. On the other hand, it must be emphasized that 
the oases could never vie with the great river plains in productivity, population, 
hence in historical importance, but they are also not to be compared with the 
primitive settlements of northern Europe. 

Finally a word must be said about the achievements of linguistics in recent years, 
which have revolutionized our view of the Iranian peoples. Not only have 
documents in several eastern Middle Iranian languages, Khwarazmian, Sogdian, 
Khotanese-Saka, and Kushano-Bactrian, revealed a rich, new east-Iranian world, but 
also studies of modern dialects have shown the complexity of the movement of tribes, 
the resettlement of villagers far from their homeland and a host of cultural problems 
in the development of the languages and dialects. As more and more data accumulates, 
our guesses about the solutions to problems become more educated, and lacunae of the 
past are gradually filled. Pioneering work in deciphering Khwarazmian was made by 
W. B. Henning, and before him F. W. K. MUller for Sogdian. 56 Khotanese-Saka was 
elucidated by the early work of Leumann, Konow and Bailey. 57 The latest Middle 
Iranian language to come to light was Kushano-Bactrian, which was studied by 
Maricq, Henning and Benveniste. 58 We now know that Khwarazmian had a long 

54 See especially the collection of articles in MUller, "Handschriften - Reste in Estrangelo- 
both Antkhnyi Corod (Moscow, 1963), and Schrift aus Turfan," Abh.PAW (Berlin, 1904). 
Drevnii Voslok, Goroda i torgoulya (3-2 mill. B.C.) 57 E. Leumann, Zur nordarischen Sprache und 
(Erevan, 1 973), for the Near East, N. Pigulevskaja, Literalur (Strasbourg, 1 91 2) ; S. Konow, in A. F. R. 
Les uilles de Vital Iranien aux tpoaues parthes et Hoernle, Manuscript Remains of Buddhist Literature 
sassanides (Paris, 1963), for Iran, and A. M. found in Eastern Turkestan (Oxford, 1916); and 
Belenitskii, ed. Srendnevekovyi gorod Srednei Azii H. W. Bailey in many articles in the BSOS and 
(Leningrad, 1973), for the Islamic period of books, beginning in 1935. 

Central Asia. 58 A. Maricq, "La grande inscription de Kan- 

55 For Central Asian archaeology see G. Frum- ishka," JA, 246 (1958), 345-440; W. B. Henning, 
kin, Archaeology in Soviet Central Asia (Leiden, "The Bactrian Inscription," BSOAS, 23 (1960), 
1970). 47-55; E. Benveniste, "Inscriptions de Bactriane," 

56 See W. B. Henning, "Ober die Sprache JA, 249 (1961), 113-52. 
der Chvarezmier," ZDMG, 90 (1936), *30-34; 



Demography 43 

history, first written in a modified Aramaic alphabet; then it was written in Arabic 
characters down to the Mongol conquests. Sogdian was written in its own alphabet, 
derived from Aramaic, or in a special Manichaean alphabet, or in a Syriac script, resp. 
in texts of three religions: Buddhism, Manichaeism and Christianity. Kushano- 
Bactrian, written in a modifed Greek alphabet, survived into the Islamic period, 
when it was replaced in writing by new Persian in Afghanistan or by Indian tongues 
to the east. 

All contemporary Iranian languages and dialects are either unwritten, or have 
recorded literatures only from the nineteenth century, save Mazandarani which was 
recorded in the Arabic alphabet just prior to the Mongol invasions, and Pashtu and 
Kurdish, both of which are later in being written. G. Morgenstierne classified the 
Indo-Iranian languages of the frontier areas, while many Soviet scholars, beginning 
with 1. 1. Zarubin brought the Pamir languages into an orderly classification. In the 
investigation of Ossetic, pioneer scholarly work was done by V. Miller and many, 
especially V. Abaev, have continued his work. One may say that progress in 
understanding the Iranian world in all of its aspects has been remarkable in the past 
sixty years, and the most exciting discoveries have come in the east, in Afghanistan 
and Central Asia, revealing much hitherto unknown information, and the new 
developments are the result largely of field work. 

In the twentieth century new technologies of recording, photography, 
transportation, etc., have opened areas to investigation which formerly had been 
difficult of access. The results of archaeology, anthropology and dialectology, all field 
work, have amply justified efforts, time and money put into them. Without the raw 
material from the field, cabinet scholars would not have data on which to base their 
studies. The combination of field and desk work is even more essential, in my 
opinion, than the study of both past and present in the understanding of the Iranian 
world, or any part of it. Because of both the unity and the continuity of this world, 
perhaps more than anywhere else, a Renaissance approach to Iranian Studies still has a 
value for all who venture onto the manifold and specialized branches of that world. 
This is why, in my view, the study of the pre-Islamic past of the Iranian world must 
begin with an appreciation not only of the complexities of that study, but also with a 
survey of disciplines such as anthropology and linguistics, which in other areas are of 
less importance to the reconstruction of the picture of the past. In the Iranian area all 
of this, and more, does indeed have relevance to how we seek to understand the past, 
and also it does aid us to appreciate the present and the future. 



CHAPTER III 

PRE-IRANIAN HISTORY OF THE PLATEAU 
AND CENTRAL ASIA 



Literature: The history of the Near East in the pre-Median period is well covered by the Cambridge 
Ancient History, especially vol. 2 (Cambridge, 1 973). On Iran see especially Chapter 7 by W. Hinz "Persia, 
c. 1 800-1 550 B.C.," Chapter 1 6 by R. Dyson "The Archaeological Evidence of the Second Millennium B.C. 
on the Persian Plateau," Chapter 29 by R. Labat, "Elam, c. 1600-1200 B.C.," and his Chapter 32 "Elam and 
Western Persia, c. 1 200-1 000 B.C.," plus R. A. Crossland "Immigrants from the North," Chapter 27 ; with 
bibliographies on the Hurrians. The book by I.J. Gelb, Hurrians and Subareans (Chicago, 1944) is useful as 
a general survey, but consult also D. O. Edzard "Hurriter" in Reallexikon der Assyriologie (Berlin, 1 975) sub 
Hurriter. 

The problem of the Ur-Heimat of the Indo-Europeans has a huge bibliography with much controversy, 
but a consensus has been reached that the homeland must have been in the vast steppes of South Russia and 
Siberia. Many linguists such as Brandenstein and archaeologists such as Gimbutas, believe the homeland 
can be further defined as the general lower Volga River region (cf. their articles in A. Scherer, ed., Die 
Urheimat der Indogermanen, Wege der Forschung, 166 (Darmstadt, 1968). Cf. also B. Schlerath, Die 
Indogermanen, das Problem der Expansion eines Volkes im Lichte seiner sozialen Struktur, Innsbrucker 
Beitrage zur Sprachwissenschaft, 8 (Innsbruck, 1973). Also of interest is P. Thieme, Die Heimat der 
indogermanischen Gemeinsprache, AAWM, no. 1 1 (Mainz, 1953). All of the scholars quoted, however, are 
cautious about defining a homeland. The main dispute, is between linguists and archaeologists, since the 
identification of certain types of burial (kurgans) as Indo-European, is not accepted by everyone. See R. 
Schmitt, "Proto-Indo-European Culture and Archaeology : Some Critical Remarks," JIES, 2(1 974), 279- 
88, and the reply by M. Gimbutas, 288-89. 

It is generally agreed, however, that the unseparated Indo-Iranians must have had close contacts with 
the early Finno-Ugrian peoples because of very old borrowings into the languages of the latter. Cf. A.J. 
Joki, Uralier und Indogermanen, Memoires de la Societi Finno-Ougrienne, 151 (Helsinki, 1973). Soviet 
archaeologists have come to the conclusion that the center of the Indo-Iranians should be sought near the 
southern Urals, and this is not contradicted by linguistic data. Cf. K. F. Smirnov and E. E. Kuzmina, 
Proiskhozhdenie Indoirantsev v svete noueishikh arkheologicheskikh otkrytii (Moscow, 1967), esp. 51-52. 

On the Andronova culture, the archaeological name for the presumed Indo-Iranian culture in 
Kazakhastan, cf. K. Jettmar, "Mittelasien and Sibirien in VortUrkischer Zeit," in Handbuch der Orientalistik, 
Altaistik, Geschichte Mittelasiens (Leiden, 1966), and G. Frumkin, Archaeology in Soviet Central Asia 
(Leiden, 1970), ch. 1. On the southern connections of this and other cultures see V. M. Masson, Srednyaya 
Aziya i Drevnii Vostok (Moscow, 1964), 184-85, and his Central Asia, Turkmenia before the Achaemenids, 
with V. I. Sarianidi (London, 1972), esp. 150-54. A survey of archaeological work to the west, in 
European Russia, may be found in M. Gimbutas, "Proto-Indo-European Culture: The Kurgan Culture 
during the Fifth, Fourth and Third Millennia B.C.," in C. Cardona and H. Hoenigswald, eds., Indo- 
European and Indo-Europeans (University of Pennsylvania, Philadelphia, 1970), 155-98. 

Archaeological, primarily pottery, evidence for the Indians (or Indo-Aryans) on the Iranian plateau in 
the second millennium B.C. has been assembled by R. Ghirshman in L'Iran et la migration des Indo-aryens et 
des Iraniens (Leiden, 1977), who proposes an early migration of the former from Central Asia, but a later 
one from the north Caucasus for the Iranians. The Caucasian migration of Iranians is rejected in favor of a 
Central Asian path by T. C. Young.The Iranian Migration into the Zagros," Iran, 5 (London, 1967), 1 1- 
34, and other archaeologists. No agreement seems possible, although the Central Asian route seems more 
plausible. 

With the coming of the Indians to the subcontinent and the Iranians to the plateau, the Vedas, especially 
the Rigveda, and the A vesta, join archaeology as two written sources, but their historical interpretation 
and attempts to find archaeological parallels to the texts are fraught with problems. An indispensable 
handbook to work on the A vesta is B. Schlerath's Awesta-Wdrterbuch (Wiesbaden, 1968), while the old 
work of W. Geiger, Ostiranische Kultur im Altertum (Erlangen, 1882) is still useful as an attempt to relate 
the Avesta to history and geography. For bibliographies see my The Heritage of Persia, second edition 



46 Chapter III 

(London, 1976), 303-05. Finally, a linguistic attempt to put the migration of the Indians into historical 
perspective is T. Burrow, "The Proto-Indoaryans," JRAS (1973), 121-40. 

RECONSTRUCTIONS 

A half a century ago it would have been possible to begin the history of the Iranians 
with the cuneiform records of the ninth century B.C. which give Iranian names for 
persons met by Assyrians in their campaigns to the east. It is clear that the Iranians, 
when they reached the plateau, came upon natives who were at various stages of 
culture and with different ways of life. In western Iran before the coming of the 
Iranians, the Hurrians had descended to the plains in the north, while the Elamites 
were dominant in the south. Whatever other groups existed, we may speak of the 
Hurrian (and later Urartian and Mannean) sphere in the north, and the Elamite 
(together with Kassite) realm of the south. The former is more difficult to understand, 
but we must discuss both before turning to the Iranians. 

From the third millennium B.C. Akkadian cuneiform tablets mention the Hurrians, 
first in the highlands to the east and later in northern Mesopotamia, Syria, and even 
Anatolia. Just as the 'Fertile Crescent' of Mesopotamia and Syria was the goal of 
Semitic invaders from Arabia, so it was also the field of expansion for immigrants 
from the eastern highlands, throughout history. And just as various Semites such as 
the Akkadians and then the Arameans spread over Mesopotamia, especially in the 
south, so did the Hurrians, especially in the north. If we use the term Hurrian in its 
widest sense, as a generic term for all of the peoples in northwestern Iran speaking 
related languages, there are three states which are recorded in history, the Mitanni 
confederacy, which flourished in northern Mesopotamia in the middle of the second 
millennium B.C., and the states of Urartu and of the Manneans, both of which flourished 
in the first part of the first millennium B.C. Other peoples such as the Lullubi and the 
Guti or Quti, mentioned in earlier cuneiform records, surely continued to exist in 
Kurdistan of today, but they seemingly had little or no political influence. The 
Mitanni are important as the Hurrian kingdom in northern Mesopotamia where the 
first Indo-Iranian names appear in history. Much has been written about the 
undoubtedly small group of Indo-European speakers who appeared from the east.just 
as other early Indo-European speaking peoples, the Hittites, Luvians, and others appear 
in Anatolia, probably at an earlier period. Much has been written also about the 
names of the deities in a treaty between a Hittite ruler Suppiluliumas I and a Mitanni 
king Kurtiwaza, which deities now everyone agrees are to be identified with Indian 
Indra, Mitra-Varuna and the Nasatyas. Likewise the terms for the training of chariot 
horses by a Mitanni called Kikkuli, found in a manual written in cuneiform Hittite, 
point also to a language used by some of the Mitanni which can be called an archaic 
Indian dialect of the Indo-European family. 1 " Other words in a Kassite glossary and 
Hurrian cuneiform texts from a site called Nuzi, attest to the presence of Indo- 
European speakers in the Near East in this early period. It is not the intention here to 
discuss any of the words, whose etymologies are much disputed, but to draw 

u M. Mayrhofer, Die Indo-Arier im alten Vordera- Vorderen Orient - Ein Mythos?, SWA W, (Vienna, 
sie n (Wiesbaden, 1 966), and his sequel Die Arier im 1 974), 294. 



Pre-Iranian History of the Plateau and Central Asia 41 

historical conclusions from the undeniable presence of Indo-European speakers in 
Mesopotamia who came from the east. 

From the outset, it can be accepted now that those Indo-European speakers were 
Indians, yroto-Indians, or Indo- Aryans (to be distinguished from Dravidians or other 
Indians), whatever one may wish to call them; they were neither Iranians nor 
unseparated Indo-Iranians. This general opinion is shared by scholars who hotly 
dispute the etymologies of various words such as Kammenhuber, Dyakonov, and 
Mayrhofer (cf. bibliographies of works in Mayrhofer, Die Arier, supra). The second 
point, that these Indians were not a group who came westward from India but rather 
an offshoot of probably the first migration of Indo-European speakers from Central 
Asia into eastern Iran or Afghanistan, is also generally accepted by most scholars. 
Further conclusions are rarely drawn, except the obvious one that these Indians, even 
though they may have been a ruling class of the Mitanni, were soon absorbed into the 
local populations of Mesopotamia and Syria. Unfortunately, we do not have the data 
to come to any more conclusions, but the important question one would ask is, what 
is the story of these Indians in Iran? Either they were only a small spearhead of a tribe 
which wandered from the east across Iran to Mesopotamia, or they were only that 
group of Indians who moved away from other Indians on the Iranian plateau down to 
the plains of Mesopotamia. If the former were true, the Indians must have been an 
unusual military band with a superior technology of war chariots which enabled 
them to cut through other peoples in Iran to penetrate like a mobile army to the plains 
of Mesopotamia. If the Indians of Mitanni were only part of a much larger group of 
Indians on the Iranian plateau, one would expect some traces of them in Iran. Also, in 
either case, the split between the Indians and Iranians must have occurred much 
earlier than hitherto supposed, and presumably in Central Asia not in Iran or 
Afghanistan. These two questions are related to each other, and we must examine 
both of them. 

The latter problem has been much discussed by both archaeologists and linguists 
because it is part of the Indo-European problem about which there is an enormous 
literature. Here too there is now a consensus that the homeland of the Indo-European 
peoples was in south Russia, while the Indo-Iranian branch was located in the 
northeastern part of that general homeland. This is, of course, all hypothetical, but it is 
the most likely theory. Also, the Indo-Iranians probably were the last, or one of the 
last, of the Indo-European peoples to move from the homeland, and the movement 
was to the south and east. The difficulty of attaching archaeological finds to linguistic 
data plagues everyone seeking to reconstruct the movement of the Indo-Iranians, but 
a general acceptable hypothesis is that the separation of the Indians from the Iranians 
took place in Central Asia, or even more to the north, and the Indians moved 
southward in the first half of the second millennium B.C. This admittedly is imprecise, 
but it is not possible to go into further details given the state of knowledge at present. 
At the same time one must account for the existence of the Dardic languages today 
and other facts, so an attempt at explanation should be made, always remembering 
that we are not dealing with factual data but surmises. 

Ghirshman reconstructs the movement of the Indians as follows 1 : the place from 

1 Pp. cil. [ch. 2, n. 32], 18. 



48 Chapter III 

which they dispersed was Gurgan and Turkmenistan where they had settled in the 
third millennium B.C. coming from the north. He bases the settlement of Indians on 
the presence of grey or black pottery which replaces earlier painted pottery, 
accompanied by clay idols, highly stylized "violin-shaped'' female figurines. Together 
with these two features, chariots drawn by horses and small trumpets in gold or silver 
complete the list of components identifying the Indians (or Indo- Aryans). Where 
these features occur in excavations, according to the author, we should expect to find 
Indians. These criteria force the author to bring the Indians to the southeast corner of 
the Caspian Sea already at the end of the fourth millennium B.C. Such an early date is 
surprising though not impossible. The first criterion for the presence of Indians is the 
grey or black pottery. But this pottery is also associated later with Iranians at sites such 
as Hasanlu south of Lake Urmia in Azerbaijan, presumably with a Mannean 
(Mannai) population. 2 The distribution over time and space of grey or black ware 
obviously precludes the incorrect syllogism that where this ware occurs we must 
have Indians. To turn to the "violin-shaped" female figurines, a special style centered 
on Turkmenistan can be argued, but the existence of similar figurines from the Indus 
valley to Mesopotamia from ages long before the Indo-Iranians appeared also cannot 
be denied, and an extensive survey of such figurines does not support Ghirshman's 
hypothesis. 3 Horses and chariots also do not inspire confidence in his theory, since at 
Uruk from the end of the fourth millennium, the character for a four-wheeled vehicle 
is recorded in cuneiform and the domestication of the horse is also early there. 4 This 
leaves the trumpet, and the evidence brought by Ghirshman for the Indo-Aryan 
origin of this is the borrowing of it by the Egyptians, possibly from the Mitanni. In 
other words, none of Ghirshman's arguments are convincing; on the other hand all of 
them conceivably may be attributed to Indo-Aryans if this term is stretched beyond 
the usual conception of these people. 

All of this brings us back to philology and how the Indians came to the Near East. 
What happened to the Indians who did not descend to the plains of Mesopotamia as a 
component of the Mitanni? As mentioned, either the Mitanni Indians were an 
isolated band who wandered far to the west leaving no remnants behind them in Iran, 
or they were only a small part of a larger group in Iran, and if the latter where are 
their traces? If traces are to be found, they should exist in the mountain refuge areas 
where the remnants of the Indians survived while the rest were absorbed by their 
cousins, the Iranians, who came later. T. Burrow, who adheres to this view, even 
suggests that the name Urmia (for town and lake) goes back to a Sanskrit word Urmi 
"wave." 5 This is unlikely, however, since in the Urartian language a number of place 

2 See E. Porada, The Art of Ancient Iran (New Reports of the USSR delegation to the 8th 

York, 1965), 108, and R. H. Dyson, "Problems of international congress of pre- and proto-historic 

Protohistoric Iran as seen from Hasanlu," JNES, 24 sciences in Belgrade, 8-9. More detailed and with a 

(1965), 211. large bibliography is her article "Rasprostranenie 

V. M. Masson and V. I. Sarianidi, Sredneasiats- konevodstva i kulta konya v iranoyazychnykh 

kaya Terrakota epokhi bronzy (Moscow, 1973), piemen Srednei Azii," in Srednyaya Aziya, ed. by 

especially ch. 4 B. G. Gafurov and B. A. Litvinskii (Moscow, 

4 E. E. Kuzmina, Earliest Evidence of Horse 1 977), 28-52. 
Domestication and Spread of Wheeled Vehicles in 5 T. Burrow, "The Indoaryans," JRAS (1973), 

Connection with Problem of time and place of the 139 
formation of Indo-European Unity (Moscow, 1971) 



Pre-Iranian History of the Plateau and Central Asia 49 

names begin with ur-, and Urmia, being in Mannean territory, is more likely a word 
from their language than from Sanskrit. Likewise the etymology of Mazandaran as 
containing the name Indra, a Vedic deity, proposed by Burrow, does not inspire 
confidence. 6 What is needed are some names which are indisputably Indian, such as 
the Mitanni deities. Unfortunately, etymologies of names are subject to as much 
dispute among linguists as the historical significance of pottery types among 
archaeologists, and none of the names suggested as being of Indian origin, Bagdatti 
and Artasari, Bagbartu, etc. can be definitely assigned an exclusively Indian 
etymology. 7 So in the realm of linguistics we depend on impressions, which suggest 
an Indian presence on the plateau or in the mountains of Iran, but there is no data 
available to confirm these impressions. If we follow Burrow in his discussion of devs 
in Mazandaran, again we are in the realm of conjecture, for the use of the term dev, in a 
favorable sense, does not assure us that Indians were present, in opposition to the 
Iranian or Zoroastrian condemnation of the devs, and their worshippers. The devs 
could, of course, refer to non- Aryan deities and names found in the Caspian region 
later, such as Devsalar 'leader of devs,' Devdad Theodore,' etc., do not necessarily point 
to Indians or even to Iranians who did not follow the condemnation of the daevas and 
daeva worship. On the other hand, the Indians of the Mitanni people must have 
existed, so the theory of a first wave of Aryans as manifested in these Indians or proto- 
Indians, before the coming of the Iranians, is not at all far-fetched. 

The existence of the Dardic languages in the northeast of the Iranian area would 
also point to an early wave of Indo-European speakers into this area. The extreme 
difficulties of comparing contemporary languages with an ancient language of over 
three millennia ago (the Indie of the Mitanni) is obvious, and further comparison 
with the religion of some of the Dardic people (the deity Imra as Indra, and the like) 
with the Vedic deities of Varuna, MitrS, etc. can lead to no certain conclusions, but 
rather little more than a subjective intuition that the remote ancestors of the Dardic 
peoples had some relations with the Indians of the Mitanni. There is, it seems, 
no objection to the proposition that when the Iranians came upon the plateau they 
were not the first Indo-European speakers that the autochthonous population had 
encountered. Surely the Indian layer on the substratum, however, was not very 
numerous, and in places it must have been absorbed early by the natives. The only 
traces of the Indians which may have survived could be the daeva worship (not devil 
worship), primarily in the Caspian provinces, which withstood the Zoroastrian 
reform of the ancient religion of the Indo-Iranians. As we shall see, a Zoroastrian 
reform is the easiest way to explain most of the divergences between the Iranians and 
the Indians in religion. 

THE NEAR EAST AT THE TURN OF THE MILLENNIUM 

Whatever the role and importance of the Indians in the Near East had been, by the 
year 1000 b.c. the autochthonous people were organized into small states or tribes in 
western Iran, according to cuneiform sources. Elam was in decline, since it had 

S. Kiyj, Shahndme ve Mazandaran (Tehran "Weiteres zur Urgeschichte der Inder," KZ, 55 
1353/1975), 32 (1927), 75-103, especially 100. 

7 Burrow.op. of. [n. 5], 140; and P. Kretschmer, 



50 Chapter III 

suffered greatly at the hands of the Babylonians over a century previously. The 
Kassites long before 1000 B.C. had been absorbed in Mesopotamia, although in the 
Zagros mountains, present southern Luristan, they continued to live, and many of the 
'Luristan' bronzes were found in graves here. Known to the Greeks as KoaaaToi, to 
the Latin authors as Cossaeans (both with variants) and in cuneiform sources as KassQ, 
this Zagros people are mentioned as late as Alexander's conquests. To the south of 
them, also in the mountains lived the Uxii or Ov£ioi (with variants), hwz' in the 
Talmud (but applied to the plains as well as mountains), Huzaye in Syriac, and 
Khuzistan of today with its capital Ahwaz (Arabic pi. oihwz). These tribal peoples 
owed no allegiance, or only temporary, nominal fealty to Elam or other later states. In 
Persis to the south were primarily Elamites with their principal city Anshan Qiodie 
Tepe Malyan), and to the north of the Kassites apparently were various peoples who 
are not clearly identifiable in the cuneiform sources. Whether the ancient, Quti/Guti, 
or other peoples still lived in the Zagros mountains at the beginning of the first 
millennium B.C. is uncertain, but surely their descendants were there. It is clear, 
however, that no states or tribes were either dominant or expanding in this area at this 
time, but this was to change greatly with a series of events which seems to have 
occurred in the first centuries of the millennium - the expansion of Assyria and 
Urartu, and, of course, the Iranians over the plateau. 

If we only had information from the first century of the millennium, some of the 
later changes in states and peoples might be explicable, but one can only guess at the 
forces at work which changed this part of the ancient world. Assyria had expanded, 
especially to the west, under Tiglath-Pileser I (c. 1116-1090 B.C.) but after his reign 
had soon returned to the size of its homeland in northern Mesopotamia. Babylon 
likewise was in a state of decline, and the apparent reason for the decline of both 
powers was the spread of the Arameans into Mesopotamia. Many small states of these 
invaders from the desert were created, but they never united into a strong kingdom. 
Instead they gradually gave their language to the settled folk and fused with them. 
The spread of the Arameans and their language Aramaic was paralleled by the 
expansion of the Iranians, and incidentally also of a much more widespread use of iron 
and steel. 8 Whether the Iranians were carriers of this expanded metallurgy is 
doubtful, but they could have utilized it well in their expansion. 

We return to the question of the Iranians who spread from their homeland much 
later than their cousins, the Indians. The Iranians, we have mentioned, remained in 
Central Asia and continued the tradition of the Andronovo culture after the Indians 
had moved south. It is interesting to note that ancient Iranians in Central Asia, and 
presumably extending into European Russia, did have contacts with the Finno- 
Ugrian peoples, but not apparently with the Baltic speakers. 9 These observations, 
based on linguistic data, coincide with archaeological finds of the Andronovo culture 
in Kazakhastan. Since the widespread (over both space and time) black or grey pottery 
is no useful index of the presence of either Aryans or Iranians, what were some other 

See C. A. Singer, A History of Technology, obugrischen Sprachen (Budapest, 1972), esp. 33-34; 

5 vols. (Oxford, 1954-58); and R.J. Forbes, and P. Arumaa, "Bakes et iranicns," Sludi Linguis- 

Sludies in Ancient Technology (Leiden, 1955- tici in onore di Vittore Pisani (Brescia, Paideia, 

58). 1969), 73-90, esp. 89-90. 

E. Korenchy, hanische Lehnwbrter in den 



Pre-Iranian History of the Plateau and Central Asia 51 

features of the culture of the early Iranians? Just as with pottery one must be careful in 
attributing types of burial and features of material culture exclusively to Iranians, for 
they too could be borrowed or adapted by other peoples. Burials in kurgans, a large 
mound with a room for the body and burial objects, and interment of horses with the 
dead chieftain or warrior, are two features of the culture of the Andronovo people. 10 
Can one maintain that wherever these two features are found Iranians are present, or 
that they are exclusive to the Iranians? In any one of these two cases, the migration of 
Iranians through the Caucasus to northwestern Iran can be proposed, since kurgan 
burials have been found in Soviet Azerbaijan near sites such as Mingechaur and 
Kirovabad. 1 ' The use ofkurgans persisted into historic times and the dating of graves 
without inscriptions is particularly difficult, since presumably the early Iranians were 
nomads or semi-nomads who both adopted and gave ideas, making difficulties in 
entangling origins. There is no reason why Iranian bands should not have gone down 
the western shores of the Caspian Sea, as well as through Central Asia. And, whereas 
the possible traces of an Iranian migration from the north to Azerbaijan and south are 
difficult to find, in Central Asia archaeological indications for a migration are more 
promising. 

In Central Asia the archaeological trail of Iranian migrations, as of any migrations 
of tribes, has not left clear traces. Soviet archaeologists, however, have opened new 
vistas on ancient Central Asia, and although excavations at Kyuzeli-Gyr in 
Khwarazm, Afrasiyab I in Sogdia, Elken tepe in Turkmenia, Yaz II in the Merv oasis 
and Kobadian I on the north bank of the Oxus River in southern Tajikistan have 
revealed similar pottery from the immediate pre-Achaemenid period, the earlier 
centuries are not at all uniform. 1 2 All attempts to establish the material culture of the 
ancient Aryans on the basis of archaeological finds, including types of burial, metal 
objects and the like, have failed to establish an overall accepted reconstruction of the 
expansion of the ancient Iranians. One feature which may confuse archaeologists is 
that cultural objects may go the opposite direction from a nomadic invasion or a 
movement of peoples. Indeed, it is likely that many aspects of early Iranian settled 
civilization, including city or town structures, came from Mesopotamia or at least 
from south to north rather than the reverse. 13 The direction of influence is always 
difficult to discern especially in early periods of history. 

It is not possible here to discuss the many ramifications of Central Asian 
archaeology, the comparative stratigraphies of such key sites as Anau and Namazga 
tepe in Turkmenistan, and the many other excavations which have given us a picture 
of Central Asia different from the old view of the area as a wide expanse of steppe 
lands sparsely inhabited by nomads. Although deserts do occupy much of Central 
Asia, elsewhere the fertility is great, as Quintus Curtius, historian of Alexander the 
Great, noted, "the land of the Bactriani is of a manifold and varied nature. In one part 

Smirnov and Kuzmina, supra, Protskhozh- Drevnezemledelcheskaya kullura Margiany, MIA 73 

*"'*' 6. (Moscow, 1959), 218 pp. A French summary of 

For a bibliography see K. V. Trevor, Ocherki the book by Ghirshman may be found in I A, 4 

po isiorii i Kulture Kavkazskoi Albanii (Moscow, (1964), 69-84. 
1959), 36. iJCf. V. M. Masson, "Problema drevnego 

A good survey of the pottery of the various goroda i arkheologicheskie pamyatniki severnoi 

sites with historical notes is in V. M. Masson, Baktrii," Dreunyaya Baktriya (Leningrad, 1974), 8. 



52 Chapter III 

many trees and vines produce plentiful and mellow fruits, frequent streams irrigate 
the rich soil, the milder parts of which they sow with grain, the rest they leave for 
pasture for the flocks." 14 There is no evidence of any large political entity in Central 
Asia before or at the time of the expansion of the Iranians, and one would not be 
expected so early. From excavations and from the Avesta, as well as the parallel Vedas, 
however, a tentative conception of the life of the early Iranian tribes may be 
reconstructed. 

The picture of early Iranian life by Geiger still retains interest for the student even 
though some of his nineteenth century romantic conceptions or ideas about race seem 
misplaced today. 15 The ancient Iranians were both conversant with agriculture and, 
of course, with pastoralism; at least the vocabulary indicates both forms of life. 
Likewise the tribal nature of Iranian society is implied by the accounts of Herodotus 
and other Classical writers which conform to all we know about it. There are no 
indications of a developed town life among the early Iranians, and we may assume 
that they encountered relatively undeveloped urban societies, and only in some of the 
oases of Central Asia and on the Iranian plateau. 16 For a long period of time the 
Iranians probably remained in Central Asia and eastern Iran with little contact to the 
west where settled cultures did flourish. We infer this since the Avesta shows no 
evidence of any contact with the states or civilizations which existed in western Iran 
or Mesopotamia. Although the 'urban revolution,' to use an expression of Gordon 
Childe, had not occurred in the areas where the early Iranians first lived before 
spreading over the plateau, some settlements even previous to the coming of the 
Iranians did exist in Central Asia. 17 Trade and commerce, as noted above, had been 
active for many centuries if not millennia before the coming of the Iranians, but 
settlements for the production of objects were few and small. 18 In oases such as Herat 
and in Seistan small centers probably existed, but we need more archaeological work 
to give us a clear picture of their extent and nature. Anthropologists and sociologists 
have proposed many criteria for the development of urbanism, or for the distinction 
between an urban and an agricultural civilization, and models have been created to 
apply everywhere. One or more of the models may fit the case of the Iranians, but 
from the only sources we have, archaeology and the ancient religious writings, again 
the dominant pastoral society of the early Iranians is what strikes one's attention. It 
was an heroic, tribal society probably similar to the life of the Pathans or Pashtuns 
today on the Afghan-Pakistan borderlands. The exploits of warriors and sacrifices to 
the gods were two important activities we infer from the Avesta. 

Controversy has not subsided over the theories of Georges Dumezil regarding Indo- 

14 History o/Vl/e.vWer (VII, 4, 26), ed. and trans. Evolution of Urban Society (Chicago, 1966); R. 
by John C. Rolfe, 2 (London, 1962), 159. Tringham.ed., Urban Settlements (Warner Modu- 

15 W. Geiger, Ostiranische Kultur im Altertum lar Publications, Andover, Mass., 1973); R. 
(Erlangen, 1882). Braidwood and G. Willey, eds., Courses Toward 

16 See V. M. Masson, Srednyaya Aziya i Drevnii Urban Life (Chicago, 1962), and R. M. Adams, 
Vostok (Moscow, 1964), 454-56. 'The Origins of Cities," in C. C. Lamberg- 

17 V. G. Childe, "The Urban Revolution," in Karlovsky, ed., Old World Archaeology (San 
The Rise and Fall of Civilizations, ed. by C. C. Francisco, 1972), 137-43. 

Lamberg-Karlovsky (Menlo Park, California, ,8 Cf. J. A. Sabloff and C C. Lamberg- 

1974), 6-14. An immense literature exists on the Karlovsky, eds., Civilization and Trade (Albuquer- 

beginnings of urbanism; cf. P. M. Adams, The que, N. M., 1975). 



Pre-Iranian History of the Plateau and Central Asia 53 

European, and, of course, early Iranian society and religion. Indeed in the Iranian field 
scholars are sharply divided on the theory or on its relevance to an understanding of 
ancient Iranian society. Briefly Dumezil claims that Indo-European society had a 
tripartite division based on a conscious ideology found nowhere else in the world. 
Although the division of people into priests, warriors and common folk may have 
existed elsewhere, only the Indo-Europeans developed this ideology as a fundamental 
basis of both their religion and their society. 19 In this division of society, which 
reflects the division of the gods, there is a further subdivision of magical and juridical 
aspects of the three functions. Thus, among the Indians, the gods Mitra and Varuna 
were the deities of the priests, the highest function of sovereignty, and this 
corresponded to Vohu Manah and Asa in Iran, two of the AmSsa Spentas "holy 
immortals" or archangels in the Zoroastrian religion. Although the Zoroastrian 
reform of the ancient religion of the Iranians changed the roles of the ancient Indo- 
Iranian deities, Dumezil nonetheless finds a continuation of the ancient Indo- 
European ideology in Zoroastrianism with a substitution of characters for the 
original roles. Almost everyone concedes that a tripartite division of society is almost 
a natural differentiation in ancient societies everywhere, the only argument 
concerning Dumezil's theory is the conscious, overwhelming importance of this 
ideology and its ramifications for the Indo-Europeans. It would fit in well with the 
caste system of India or the division of society into three classes in Iran, but the 
significant ramifications of this are difficult to find. To say that one important result 
of the division of society was the Mannerbund or group of young warriors, upholders 
of the second function of force or military power, does not single out Indo-European 
society as greatly different from other societies. If the young men of the Indo- 
Europeans had better weapons or horses than the young men of another group, one 
can only say that in a conflict the former would win. Just how far one can use the later 
Turkification of Azerbaijan or Anatolia as a model for the earlier expansion of the 
Iranians is subject to dispute, but the process by which local people adopted the 
language of their conquerors must have had some analogies in both cases. If there had 
been a clear tripartite ideology among the Indo-Europeans including the early 
Iranians, it is nowhere proclaimed or spelled out in the sources. It may well have been 
understood by everyone thus obviating any need to explain or propound it, but it is 
surely not found in unequivocal fashion in any sources. 20 

To return to the Iranians, regardless of how profound a meaning the tripartite 
divisions of the cosmos and society had for these people, the rites and incantations used 
to propitiate or implore the deities for aid must have survived from the hoary Indo- 
European past, with possible additions or changes according to new circumstances of 
life or contact with other peoples. On the other hand, in religion conservatism is the 
rule, and an examination of the text of the A vesta does not reveal any autochthonous 
elements. This suggests that the ancient Iranian beliefs were perhaps closer to the 

The output of Dumezil is prodigious, but The New Comparative Mythology (Berkeley, 1973), 

since he insists that one read only his latest works, and in G. J. Larson, ed., Myth in Indo-European 

see his Mytheetepopk, 3 vols. (Paris, 1968-73), and Antiquity (Berkeley, 1974). 
his earlier L'idMogie tripartite des Indo-Europtens 20 Most divisions of any kind can only be dual 

(Bruxelles, 1958). An analysis of his work, or tripartite, or multiples thereof, so the peculiarity 

including critics, may be found in C. S. Littleton, of such classifications is not apparent. 



54 Chapter III 

general Indo-European religious milieu than the religions of many other daughter 
linguistic groups. By the time the Iranians spread over the plateau, however, we may 
say that the social and moral aspects of the deities in their pantheon were paramount, 
and the functions of the gods were more important than identifications with forces of 
nature. Thus Mithra (Vedic Mitra), while associated with or even identified as the 
sun, was more the god of respect for contracts and oaths among the Iranians, which 
was part of the responsibility of humans to follow or to have fta (Skt.) or ala 
(Avestan). This important concept in the religion of the Indians and the Iranians 
which has been well elucidated by Mary Boyce as the concept of order in its widest 
sense - cosmic, social - and the order of sacrifice, plus truth among men, was the basis 
of the philosophy of life of the ancient Indians and Iranians. 2 1 The pantheon of the 
Iranians presents a problem with the name of the highest god Ahura Mazda 'Lord 
Widsom,' seemingly an epithet. Although much controversy exists about the origin 
and identification of Ahura Mazda, it is generally accepted that this deity is somehow 
related to the Indie Varuna. Just how the Iranian Varuna lost his position and Ahura 
Mazda assumed the role of chief god for the Iranians is an enigma, but Boyce 
proposed that Varuna was tied to another deity of the waters Apam Napat, and 
whereas in ancient India the figure of Apam Napat receded and Varuna remained 
important, in Iran the reverse happened. Further, in Iran Ahura Mazda assumed the 
role of Varuna, while Mithra/Mitra remained important for both Iranians and 
Indians. 22 There is much discussion among scholars about the various roles of the 
deities in the pantheon of the ancient Iranians, and here is not the place to investigate 
such details, but rather the role of religion in the life of people and related problems 
concern the historian. In both India and Iran hymns were sung to the deities as part of 
the ritual or sacrifice, and the hymns were passed on from priest to priest from hoary 
antiquity. If anything, the Iranians were more universal than the Indians in their 
objects of praise or worship, for in the Avesta we find praise for plants, animals, 
mountains, rivers, the days of the year, and in fact much of creation is revered (e.g., 
Yasna 17, 17). We do not find the transposition of identities of one deity to another in 
the Avesta as in the Rigveda of India. Thus in RV I, 94, 13 Agni is identified with 
Mitra and in RV II, 1 , 3 with Indra and Visnu, but in Iran we do not find this in the 
hymns. It is difficult to know what has been changed or added in later times to the 
text of the Avesta, but such additions seem to conform to the spirit and continuity of 
the ancient religion. In fact, the "long tradition" daraya upayana is mentioned several 
times in the Avesta as something to be revered (Y. 17, 13, etc.). Repetition of 
formulae is also a constant feature of the Avesta, and the only important changes or 
additions to the ancient religion probably were made by Zoroaster. 

Before turning to the prophet, however, one may mention several features of the 
religion which seem to have been in operation before Zoroaster. One was surely 
ancestor worship, found all over the world of the Indo-Europeans, and in the Avesta a 
special Yast 13 is devoted to the spirits of the departed including animals, birds, etc. 
(verse 74). Prayers to mountains, rivers, the elements (fire, wind, water, etc.) have 
been mentioned, and we shall meet this feature again below in considering the 

21 Boyce, A History of Zoroastrianism, 1 (Leiden, 22 Ibid., 48-52. 

1975), 27. 



Pre-Iranian History of the Plateau and Central Asia 55 

cuneiform tablets in Elamite from Persepolis. Ritual purification and ablutions were 
probably also a part of the religion of the ancient Iranians, and in later times fire 
occupied a special place in the worship of Zoroastrians. The eclipse or rise of certain 
deities, the dropping of names for epithets, and the changes in functions of certain 
spirits, in both India and Iran, all have provided scholars with much material for 
speculation. Fortunately, the detailed account of the gods of pagan Iran by Boyce 
obviates the need to comment on these religious and etymological questions. 23 It 
should be noted that the demons and evil forces in the world were as innumerable as 
the good spirits, and it seems that at times propitiation of demons in many forms was 
as much a part of some rituals as were the hymns and offerings to the benevolent 
spirits. Also noteworthy is that legends in the Avesta about ancient heroes blend with 
those of the gods. Obviously all things in life, health and sickness, poverty and wealth, 
and many more, were bound to religion in words and deeds, hymns and sacrifices. 
Belief in an afterlife also was a feature of ancient Iranian religion, but burial customs as 
recovered by archaeologists are notoriously difficult to interpret as reflecting 
theological or eschatological beliefs of those who practiced them. Whether the rulers 
or aristocracy espoused different means of disposal of a corpse than those of common 
folk is not easy to determine, but archaeology indicates that many of the early Iranians 
buried their dead in the ground. 24 In this connection the ancient rite of the horse 
sacrifice (RV I, 162) the aivamedha, the skeletons of which have been found by 
archaeologists in graves in south Russia in later Scythian times, indicates the 
importance of burial for at least kings and warriors. For both Indians and Iranians 
horses and chariots were important, as. we find in the Vedas and Avesta (Druaspa Ya$t 
I, 2; Mithra Yak, 10-11, etc.). Furthermore, the horse was a symbol or incarnation 
(avatar) of various deities, including Tistriya, the star Sirius (Yast, 8, 46). It was a 
feature of both Indian and Iranian religious belief that epithets and symbols were 
interchangeable among the deities, and thus a picture of a universal or even 
syncretistic pantheon of deities, overlapping each other's functions and domain, is 
revealed. This should have enabled the Iranians to integrate local spirits or deities of 
other peoples, such as those of the Elamites, into the vast Iranian pantheon. Probably 
only later did real syncretism become fashionable, when the gods of the Greeks were 
identified with Iranian deities. Obviously what we have preserved of the ancient 
Iranian and Indian religions is the priestly traditions, what one might call scholastic 
traditions, reflecting the beliefs, rituals and legalities of one group over the millennia. 
The mythology of the Avesta may well have been understood differently by priests 
than by the common folk, who were surely more inclined to magical interpretations 
of hymns, special incantations to ward off evil, and less formal interpretations of the 
meanings of sacrifices and litanies than those of the priests. The cult of haoma (Skt. 
soma) which produced ecstasy in the participants who drank the draught prepared 
from the plant, whatever it was, necromancy, divination and other popular 
manifestations of the supernatural, were probably more important for the common 
folk than the elaborate hymns to the gods and the rituals which were the specialty of 

Ibid., 22-84. of graves in those places and times when Iranians 

It is difficult to assign certain graves to were living indicates that some Iranians buried 

Iranians, and to distinguish between nomadic and their dead. 

settled burial customs. Nonetheless, the existence 



56 Chapter III 

priests. The Avesta, as we have it, in general shows the result of much priestly 
preoccupation with the text, and rather than devote more space to continued 
discussion of the religious beliefs of the ancient Iranians, which have exercised 
generations of scholars in Iranian matters, let us turn to the society and perhaps 
political order indicated by the religious texts. 

From the absence of words and expressions for city or even state in the Avesta, one 
may conclude that early Iranians lived in a tribal society based on the extended 
family. In both the RV (e.g., I, 96, 3) and the Avesta (Yast 8, 56; Yast 13, 10) the 
Aryan tribes or Aryan lands are mentioned, indicating a strong sense of solidarity and 
difference from the subjugated peoples. The family (Av. nmana) and clan or extended 
family (vis) were the basis of authority in ancient Iranian society with the pater 
familias (nmanOpaiti) under the vtspaiti, the clan leader. The zantupaiti, the tribal chief, 
on the other hand, seems to have represented more than just a larger unit of people, 
but was also a chief of a district or the land where the tribe had its home. It is difficult, 
of course, to distinguish between territorial and human units, but the next higher 
division of society, the dahyu, is apparently a geographical more than a social division, 
although if one translated this word as 'people' rather than more correct 'country,' the 
hierarchy in society would be maintained. Beyond this stratification an even higher 
confederacy of 'peoples' is implied in some parts of the Avesta, notably in Yast 10 
dedicated to Mithra, where a chief of some such confederacy is mentioned (verse 18: 
fratemadat, and verse 115). Although some scholars have claimed that this means the 
existence of an empire, I. Gershevitch has convincingly argued that we should rather 
understand a confederation of Iranian peoples, perhaps headed by a council with a 
chief of that council. 25 Perhaps the easiest way to indicate the old Iranian hierarchy of 
groups is to look at the Old Persian cuneiform inscriptions where the family or tauma 
(Av. nmana) is the narrow family of Darius, and the clan, or extended family, would 
be viO, here Achaemenid (Hakhamanishya) ; the tribe (unattested in OP) would be the 
Pasargadae, and finally the people or nation (OP dahyu), in the case of Darius, were the 
Parsa or Persians inhabiting Persis or Fars. Above this in the case of the Achaemenids 
was an empire instead of a confederacy as previously in eastern Iran. The Avestan 
words sastar, hamo-xhdra, xlaya, and others, seem to be epithets of the ruler of the 
dahyu, or possibly of a chief of the rulers in a confederacy. Later, in western Iran, of 
course, different concepts of rule developed with the Achaemenid Empire, which will 
be discussed below. 

In addition to the political hierarchy above, there existed the social division already 
mentioned, between priests, warriors and common folk (Yasna 40, 3). Caste divisions, 
as known in India (Skt. varna), are not found in Iran, and the special position of the 
priests later, an exclusive and hereditary situation, was surely the end result of priestly 
endeavors to exalt their own standing. As in most tribal societies, presumably also 
among the ancient Iranians, the young warriors of the tribe were of prime 
importance. Much has been written about the Old Persian word marlka, usually 
translated as 'menial' or 'servant,' although the interpretation of this as the young 

25 1. Gershevitch, The Avestan Hymn to Mithra synonym, cannot be discussed here, but the picture 
(Cambridge, 1959), 296-99. The relation of the of the hierarchy is unchanged. 
Gathic term SoiOra (Skt. ksftra) to zantu, possibly a 



Pre-Iranian History of the Plateau and Central Asia 51 

warrior of a tribe makes more sense. 26 Whether the young Iranian warriors followed 
special cults apart from the rest of the people, as suggested by Wikander, is possible, 
but any definite information about them is lacking. 27 Priests, on the other hand, 
dominate the scene in the religious writings which have survived, again probably a 
distortion of their real status in the society of the time. There are many words 
denoting priests but one of special concern to us is kavi, a word with a long tradition 
in Iranian languages, but which originally probably meant a seer or wise man as in 
India. 28 Somehow and somewhere in eastern Iran the kavis seem to have attained 
temporal power and started dynasties of rulers in small areas. At least this is the easiest 
interpretation of the title at the time of the prophet Zoroaster, to whom we must now 
turn. 

The figure of Zoroaster is at the core of the Avesta but his life and times are 
anything but historical, since legends have accumulated about him throughout the 
ages. Most scholars now agree that the place of Zoroaster is in eastern Iran because of 
the language of the Avesta and the geographical knowledge found there, both of 
which refer to the east. To narrow the geographical limits further is difficult, but the 
view of Henning that Herat and Merv was the most likely candidate for the area of 
activity of the patron of the prophet Kavi Vishtaspa has much to commend it. 29 
Because of the shifting of place names by the movement of people any geographical 
identifications must be subject to great scrutiny, and no real determination of the 
place of Zoroaster's activity is possible. The closeness of the Khwarazmian language of 
later times with the two dialects of the Avesta, Gathic and the dialect of the rest of the 
Avesta, was shown by Henning, although he cautioned that Khwarazmian cannot be 
considered a descendant of Avestan. 30 The movement of the Iranian Sakas in the first 
centuries B.C. to Seistan probably contributed to legends about Zoroaster's activities in 
their new home, while those traditions assigning the prophet to Azerbaijan are even 
later in origin. Thus we can only say that somewhere near the Herat area would be a 
likely place for the area where Kavi Vishtaspa flourished. 

The date of Zoroaster has produced much controversy, some scholars such as 
Herzfeld assigning the prophet to the time of Cyrus with the assumption that 
Vishtaspa is to be identified with the father of Darius who bore the same name. 31 
Others adopted the tradition of some later Zoroastrian writings that the prophet lived 
258 years before Alexander," a tradition whose reliability was ably defended by 
Henning. 32 His principal argument was the likelihood that any fabricated number 
would have some magical or other, perhaps historical, overtones, whereas 258 had 
none and therefore should be considered reliable. This date was accepted as genuine in 
Sasanian times and was reported in the later Arabic writings of al-BTrunT, Mas 'udland 
others, even though the interpretation of "before Alexander" was much disputed. 33 

Cf. Frye, supra. The Heritage of Persia, 55. of articles, ed. by B. Schlerath, Zaralhustra 

27 S. Wikander, Der Arische Mannerbund (Lund, (Darmstadt, 1 970). 
19 ^8). 30 Henning, op. cit., 44-45. 

On the term see Boyce, op. cit. [n. 21 ], 1 1 , and 3 ' E. Herzfeld, Zoroaster and his World (Prince- 

^"shevitch, op. cit. [n. 25], 185. ton, 1947), esp. 161 and 204. 
W - B - Henning, Zoroaster (Oxford, 1951), " Op. cit. [n. 29], 40-42. 

43. His name probably means 'one who can handle "See S. Shahbazi, "The 'traditional date of 

camels well', an appropriate name for eastern Iran. Zoroaster' explained," BSOAS, 40 (1977), 29- 

On the prophet and his message, see the collection 30. 



58 Chapter III 

Among some scholars, however, the origin or significance of the number which was 
passed to posterity was disputed, and Shahbazi suggested that the date "258 years 
before Alexander" was taken by Zoroastrian priests from Babylonian sources, which 
would have played down the reigns of the early Achaemenid kings, Cyrus, Darius 
and Xerxes, who conquered or suppressed uprisings in Babylonia, whereas Artaxerxes 
I, who had several Babylonian wives, does appear in the Zoroastrian tradition. 34 On 
the basis of a Babylonian origin for the year 258, Shahbazi finds that 539 B.C., the date 
of Cyrus' conquest of Babylon, is 228 years before the Seleucid era, plus 30 years for 
the prophet's life before he received a call to propagate the good religion. The 
Babylonian origin for the dates of early Zoroastrian tradition is an interesting 
suggestion which at the least supports an older date for Zoroaster. To go further and 
date Zoroaster at 1080 B.C., on the basis of a statement of Xanthos of Lydia and the 
Khwarazmian calendar is perhaps overly bold, but the traditional date has now been 
seriously challenged. 35 Inasmuch as centuries are but as a day for prehistoric times, it is 
hardly possible to assign even a century for Zoroaster's activity, and Boyce's "about 
1 500 B.C., or even earlier," places perhaps too much weight on the hoary antiquity of 
both the Gathic dialect and on archaic rites and the contents of the hymns. 36 To give a 
date to Zoroaster, even in a certain century, is hazardous. 

Since Zoroaster is probably to be dated in prehistoric times and placed in eastern 
Iran, the continuity of his teachings, the basis of the religion we call Zoroastrianism, 
throughout so many centuries to the present is most striking. This is not to deny 
changes in the religion, but his personality and teachings must have made a powerful 
impact on his contemporaries to have survived. Many scholars have suggested that 
Zoroaster was a traditional priest who broke with the ancient religion and became a 
monotheist. There is nothing, however, to suggest that the prophet did more than 
inject his own moral and ethical ideas into the ancient religion of the Iranians seeking 
thus to reform it. At most he could be characterized as advocating monolatry, since in 
the Gathas he appeals to Ahura Mazda, although other deities are assumed to be 
helpers of the great god, while followers of the evil spirit Angra Mainyu (MP 
Ahriman) are not denied existence by the prophet. This is not the place to discuss the 
doctrine of the six AmSsa Spentas, helpers of the creator Ahura Mazda, or of the 
ethical dualism and the choice of man, but we must seek to recover the historical 
significance of the Zoroastrian reform of the old religion. From the Avesta it is clear 
that Zoroaster had no easy time propagating his ideas. The turning point came when 
Kavi Vishtaspa accepted his teachings, but this led to many conflicts which are 
indicated in the Yasts. Possibly one reason for the slow propagation of the faith was 
that Vishtaspa seems to have been the last of his dynasty. Perhaps the initial 
establishment of the faith by the sword was followed by a more pacific spread by 
priests, since Yasna 42, 6 honors "the return of the priests (athrauan), who go (to those) 
afar who seek asa (truth, righteousness) in (other) lands." The FarvardTn Yast 13, 94 
speaks of the spread of the Mazdayasnian (Mazda-worshipping) religion all over the 

u Ibid., 32 if M. Boyce, A Persian Stronghold of Zoroas- 

35 Ibid, 34 See also Boyce, op cit. |n 21], 17, trianism (Oxford, 1977), 17 

189-91, with other arguments for a very early 

date. 



Pre-Iranian History of the Plateau and Central Asia 59 

world, and the many personal names, as well as place names, must have been real, 
although they cannot be identified. Verse 143 speaks of the fravahs or spirits of asa 
believing men and women of the Aryan, Tuira, Sairima, Sainu, Danu and other lands 
which are honored in prayer. The persons mentioned as belonging to these groups or 
lands have Iranian names yet there must have been something to separate them, since 
only the first are particularly mentioned as the Aryans. In the Iranian epic tradition 
the Toira, Sairima, and Sainu were identified as descendants of three brothers Erech, 
Salm and Tuc, Iranian counterparts of Biblical Shem, Ham and Japheth. The Sairima 
have been identified as Sauromatae in South Russia by Marquart, while the D5nu 
with the Dahae of Classical sources to the east of the Caspian Sea. 37 The Saini are 
enigmatic while the Toira or Tura present many problems, since they became known 
as the enemies of the Iranians par excellence in the epic tradition, and with the advent of 
the Turks in this part of the world they were identified with the ancient Tur people, 
who may have been originally an Iranian tribe which actively combated Zoroaster's 
teachings. The district of Turan (Kalat in present day Baluchistan) may reflect a 
movement of the Tura to the south, but it is of interest that it is today occupied by 
Dravidian-speaking Brahuis, who seem to have fallen into the same category as the 
Turks to the north, both enemies of the Iranians. 

All attempts to relate the Avesta, or the ancient Iranian epic tradition as found in 
the Shahname of Firdosi, to historical events in eastern Iran have failed, since there is 
nothing other than the Avesta itself as a source. Poetic fiction and tales are so mixed 
with what might have been fact in the Avesta that it is not possible to distinguish 
them from each other, but Mary Boyce is probably correct in saying that "both 
secular and priestly traditions transmitted by minstrel poets as well as by religious 
schools," mixed with superstition and tales are found, and this mixture has been 
transmitted down to the present. 38 

We cannot dwell on the basic features of Zoroastrianism, perhaps easily 
characterized as a threefold division of ethics (speaking, thinking and doing good), the 
correct performance of acts of sacrifice, and the keeping of the laws of purity, all of 
which will help the individual to secure salvation in the world to come. The cosmic 
history, or the "world year" of the Zoroastrians, however, should be mentioned since 
it does figure in problems of chronology connected with the Zoroastrian sources. 

The origins of the "world year" divided into periods of a millennium each are 
uncertain, but Babylonian preoccupation with astronomy, mathematics and 
chronology suggest the lowlands as a possible place of origin. 39 Some Iranian sources 
give the number of millennia as nine, others as twelve, but there are further 
disagreements on the significance of the various periods of three thousand years each. 
The first period is said to be creation in the spiritual (menog) state, followed by the 
second period of creation of the physical universe (gitig), and in the third period 
Ahriman launches an attack and brings sickness and death, but at the end of this period 
of strife Zoroaster is born, and he receives his revelation at the very beginning of the 



31 J 

38 [ 

39 - 



J- Marquart, ErclnSdhr (Berlin, 1901), 155. late development in time speculation, but such 

[n- 36], 108. ideas were prevalent. Cf. "Berossos," by C. F. 

The exact form of the Zoroastrian 'world Lehmann-Haupt, Realkxikon der Assyriologie, 2 

year' is not attested in Babylonia and it may be a (Berlin 1932-), 9-11. 



60 Chapter III 

fourth period. 40 Again it is not possible to follow the variations in prophecies about 
the last period of existence, but it had a cyclic character with good and evil times. The 
idea of one savior and then another, one every millennium, also developed in 
Zoroastrianism, together with apocalyptic ideas and the final judgment when good 
conquers evil. Beliefs in the world periods not only influenced ideas of chronology, 
but later prophets of doom or salvation appeared throughout the history of Iran 
exercising influences on the common folk. 

It should be remembered that we are talking about prehistory when history is only 
latent, or a history to be induced from few traces left by peoples, and those often of an 
enigmatic interpretation. Our reconstructions are based on surmises, and it is thus 
necessarily an impressionistic view which is presented. Any kind of quantitative 
history or 'manifest' history based on specific historical sources is impossible and 
probably always will remain so. The added accumulation of potsherds or other 
objects of material culture can only go so far in aiding our reconstruction of the past, 
and to reconstruct history from the Avesta is more like using the Book of Psalms 
rather than the historical books of the Old Testament as a source for recreating a 
history of the Israelites. Even the threefold division of society for that ancient period 
must be inferred, since there is no explicit statement of this. We can make some 
guesses about the early history of Zoroastrianism, but they remain only surmises. 

Although the Gathas of Zoroaster usually speak in generalities about good and evil, 
Yasna 46 is more specific about the prophet himself: 'To what land to flee, where to 
flee shall I go? They keep (me) away from my own (family) and my friends. Neither 
the community has satisfied me nor the deceitful rulers of the land." He continues to 
mention those who helped him in his need, Fry3na the Tura and Kavi Vishtaspa. This 
information has been amplified by later traditions which tell of the tribulations of the 
prophet and struggles to establish acceptance of his teachings. Just how much can be 
read into the early struggles of the faith, a conflict between settled herders and 
agriculturists who followed Zoroaster, and nomads who rejected him, is impossible 
to determine, but conflict there was. On an ideological or religious basis, it was 
probably primarily between the Zoroastrians who worshipped Ahura Mazda with 
his Amasa Spentas and the old daeva (Indian deva) worshippers, for whether it was 
Zoroaster who initiated the opposition between good ahuras and evil daevas, or that 
he only emphasized what already existed, this opposition has been a fundamental 
difference between the Indians and Iranians. Surely some Iranians did not follow 
Zoroaster's reform but rather opposed him, since daeva worshippers are mentioned in 
later sources, including the Old Persian inscriptions. Since the history of Zoro- 
astrianism has been recounted in great detail by Boyce, it is unnecessary to repeat 
controversies over interpretations of details of the religion. Suffice it to say that for 
centuries the Zoroastrian faith had to combat opposition and in the process it codified 
its own rites, rituals and beliefs into a way of life which held the Zoroastrians together 
down to the present. 



40 The MP texts about the world era are 
translated in and discussed by R. Zaehner, Zurvan 
(Oxford, 1955), 95-100. 



Pre-Iranian History of the Plateau and Central Asia 61 

ARCHAEOLOGICAL AND EXTRA-AVESTAN EVIDENCE 

A passage in Herodotus (III, 117) says that a certain plain in Asia surrounded by a ring 
of hills in which five passes may be found, used to belong to the Khwarazmians, but 
now to the Achaemenids. Further, that plain was located on the boundaries of the 
Khwarazmians, Hyrcanians, Parthians, Sarangians and Thamanians. The last but one 
may be identified with the people of Drangiana (medieval Zranka) and the last with 
people living to the east of Seistan. From the words "used to belong to the 
Khwarazmians," scholars beginning with Marquart have deduced the existence of a 
large Khwarazmian state. 41 There is no tangible evidence, however, for the greater 
Khwarazmian empire, and the further supposition that Herodotus refers to an 
extensive irrigation system in the plain, thus requiring the existence of a centralized 
state, is unwarranted. It is true that archaeologists have uncovered ancient settlements 
on the Murghab and Tejend Rivers, but again no evidence for a large state is 
available. 42 The results of archaeological investigations show rather the absence of 
urban centers, which would not suggest an ancient centralized state in the east. In 
Khwarazm itself archaeological work has revealed the existence of fortified living 
quarters or even villages from very early periods, but they tell us nothing of the 
political situation in that area. 43 On the other hand, many scholars believe that the 
ancient homeland of the Iranians, called Airyandm vaejah (Iranian region), is to be 
identified with Khwarazm, which is in accord with lists of place names in the 
Vendidad 1, 2 (see also Yast 5, 17, etc.). 44 While this may be true, as the natural 
geographical halting place of tribes moving from the north, the place of Khwarazm 
in the Zoroastrian tradition is unclear, for we cannot determine later changes made by 
priests in the tradition. 

The various Iranian tribes must have dispersed from the land north of the Oxus 
River to the lands where we hear of them later, not all, however, moving at the same 
time. To the east went the Bactrians, probably the easternmost group of Iranians at 
this time. They may have displaced older Indo-Iranian or Dardic peoples living on the 
plains of modern Afghan Turkestan, such that remnants of these earlier peoples are 
today found only in the mountains. The legends associating Zoroaster with Balkh 
(ancient Baxtri-) seem to be late in appearance and have no discernible historical basis. 
The name is derived from the river Bactrus (from Pliny, Nat. hist. VI, 48, 52) which 
means 'the bestower, distributer' according to Marquart. 45 Both the origin and the 
etymology of the name, however, are open to question, since it is uncertain whether 
the name is parallel to other Old Iranian tribal names such as Mada or Parsa, or if it is 
an autochthonous geographical name. If it is in fact originally a river name, it would 
parallel Harahuvati or Arachosia and Haraiva or Herat. The existence of a large 

41 J. Markwart, Wehrot und Arang (Leiden, 44 Marquart, EranSahr [n. 37], 118, 155, and E. 

1 938), 8-1 2, and in other writings. Also Henning, Benveniste, "L'ErSn-Vel et rbrigine legendaire des 

op. at. [n. 29], 42. Iraniens," BSOS, 7 (1934), 265-74, and A. 

V. I. Sarianidi, "Pechati-Amuleti Murgabs- Christensen, Le premier chapitre du Vendidad et 

kogo stilya," SA, 1 (1976), 42-68, with further I'histoire primitive des tribus iraniennes (Copen- 

bibliographical references. hagen, 1943), 66-67. 

43 S. P. Tolstov, Aufden Spuren der altchoresmi- 45 J. Markwart, A Catalogue of the Provincial 

schen Kultur (Berlin, 1953), esp. 103-12. Capitals of Eranshahr (Rome, 1931), 34. 



62 Chapter III 

Bactrian state before the Achaemenids also has been surmised by many scholars, but 
this assumption is even more tenuous than that of a Khwarazmian empire. Whatever 
the political realities in eastern Iran and Central Asia, we may conclude that the 
various oases must each have had a political organization based on the clan or tribal 
structure of the Iranians, but beyond a possible confederation of such oasis states we 
are not justified to go, at least on the basis of the scant source material which exists. 

If we assume the existence of one or more confederations of oases with princelings, 
another question to ask is whether there was a solidarity on inner tribal lines only, or 
whether supra-tribal groups existed. That is, were the Bactrians only united among 
themselves in some kind of a Bactrian kingdom, or did the Bactrians and Sogdians, or 
others, have a large confederation of several tribes? Again, there is no evidence which 
can provide a satisfactory answer. If we remember, however, that the confederation 
of the Medes first came into being only at the end of the seventh century B.C., most 
likely because of the threat of Assyrian conquest, the existence of a parallel eastern 
confederation seems unlikely. The later satrapy lists of the Achaemenids cannot be 
used profitably for a reconstruction of past combinations of lands or peoples. In short, 
we cannot find evidence for the existence of any large state or even confederacy in 
eastern Iran or Central Asia before the Median and Achaemenid empires. 

We have mentioned that Harahuvati (Skt. SarasvatT), the name of a river probably 
the present Arghandab and/or the Helmand, gave its name to the district of 
Arachosia, present Qandahar. In the early first millennium B.C. Iranians probably 
were mixing with Indians and pre-Indo-European peoples in this border area with 
India. Around the lake where the Helmand River empties, Iranians also must have 
mixed with a settled autochthonous population which had been engaged in trade and 
agriculture long before the coming of the Iranians. Archaeological excavations at 
Shahr-e Sukhte, and other sites in Seistan, confirm the antiquity of settlements and of 
extensive trade relations at an early period. 46 We can only speculate on the process 
whereby Iranians spread over Seistan, which, we remember, had the name Zranka in 
Old Persian, possibly the name given to the unusual mountain plateau in the middle 
of the lake today called Kuh-e Khwaja. 47 It should be noted that the names of Iranian 
peoples found in the Old Persian inscriptions are a mixture of tribal and geographical 
names, and do not represent a catalogue of various Iranian tribes who settled down 
after their migrations. Also it is difficult to determine whether names of peoples 
found in Classical sources from this part of the Middle East are Iranian in origin, or 
non-Iranian, and many explanations of names such as Derbikes, a tribe east of the 
Caspian Sea, the Tapuri, in the Elburz mts., and others, do not inspire confidence. 
Historically one would expect Iranian and autochthonous peoples to co-exist for long 
periods of time in any area until the Iranians eventually became dominant, at least in 
giving their language to others. 

Archaeology is slowly filling gaps in the picture of the material culture of the 
ancient Iranians, but problems of connecting the discovered material culture with 
Iranian-speaking peoples are many and complex. There is as yet no consensus on those 

46 M. Tosi, "Excavations at Shahr-i Sokhta," 47 G. Morgenstierne, "Notes on Balochi Ety- 

Iran, 10 (1972), 174-75; 11 (1973), 204, and mology," NTS, 5 (1933), 43. 
succeeding numbers. 



Pre-Iranian History of the Plateau and Central Asia 63 

aspects of life which can be called distinctively Iranian. Whether the future will soon 
see a breakthrough in this regard remains to be seen. 

In conclusion, the exceedingly scant source materials from eastern Iran and Central 
Asia, with a consequent general reconstruction based primarily on archaeology and 
stories in the Avesta and later Zoroastrian traditions, plus the epic, give us a hazy and 
very impressionistic picture of the pre-Achaemenid history of the entire area. It is 
only with the formation of the Median state that we can say that pre-history passes to 
history. 



CHAPTER IV 
MEDES, SCYTHIANS AND EASTERN RULERS 



Literature: Little archaeological work has been done on Median sites, and the written sources are most 
scanty. The literary sources have been assembled by I. M. Dyakonov in Istoriya Midii (Moscow- 
Leningrad, 1956), 485 pp. while I. Aliev in his Istoriya Midii (Baku, 1960), 360 pp. concentrates on 
archaeology, and especially the ethnogenesis of the Medes. Since the appearance of these two works, both 
of which have ample bibliographies of former publications, work on the Medes has been conspicuous by 
its absence. The popular work by W. Cullican, The Medes and Persians (London, 1965), 190 pp. adds 
nothing and emphasizes art and architecture. Archaeology is still all-important for the early history of the 
Medes and several sites are presumably Median, notably Tepe Nosh-e-jsn near the modern town of 
Malayer south of Hamadan. For a series of reports on succeeding yearly excavations headed by D. 
Stronach, see Iran, 16 (1978), 1, note 2, for a bibliography; the date when the fort and temple complex 
flourished was established from about 700 to 600 B.c. Unfortunately no inscriptions have been found 
which would give us information about the settlement, and we must rely on the material remains to 
reconstruct a picture of this Median settlement. Two other sites with Median strata are BsbJjSn Tepe near 
Harsin to the west of NQsh-e J5n in Luristan and Godin Tepe near Kangavar. For the former see C. GofFin 
Iran, 16 (1978), 29, note 1 , where a bibliography of former articles on the site is given, and for the latter 
see T. C. Young and L. D. Levine, The Godin Project: Second Progress Report, Royal Ontario Museum of 
Art and Archaeology, Occasional Papers No. 27 (Toronto, 1974). The discovery in 1967 of a large palace 
with a columned hall at Godin Tepe, together with the fort at Nosh-e J5n, indicated both the monumental 
tastes of Median rulers of this period and their quest for security from enemies. Comparisons of 
architectural features such as the narrow windows or 'arrowhead' slots in the wall show the widespread 
use of common motifs in western Iran, and the central position of the Medes in the give-and-take of such 
features. It is, of course, in the present state of knowledge, and before excavations at Hamadan, not possible 
to assign to the Medes any role of innovation in such features. Suffice it to say that the Medes participated 
in the general cultural milieu of the eighth century B.C. which seems to have been dominated not only by 
the political power of Assyria but by Mesopotamian culture. 

Assyrian cuneiform sources do not give much direct information about the Medes, for only in the 
campaigns of the Assyrian kings to the east do we find mention of the Medes, first in the time of 
Shalmaneser III (Salmanasar) about 835 B.C., on which see Die Welt des Orients, 2 (1955), 156. The 
geography of Media in ancient times has been studied by L. Levine, "Geographical Studies in the Neo- 
Assyrian Zagros," Iran, 11 (1973), 1-27 and 12 (1974), 99-124, and he convincingly argues that the 
horizons of Assyrian conquests must be greatly limited, since they did not at any time cross the Alvand 
mountain range west of Hamadan. References to texts and translations of Assyrian texts relevant to the 
Medes may be found in the footnotes to Levine's article, especially pp. 1 06-1 1 9. To the publications of the 
Ancient Records of Assyria and Babylonia, texts and translations by D. D. Luckenbill (Chicago, 1926-27), 
and the Prisms of Esarhaddon and of Ashurbanipal by R. Campbell Thompson (London, 1931), add the 
articles of E. Michel on "Die Assur-Texte Salmanassars III.," in Die Welt des Orients, 1 (1947), 67 foil., and 
D.J. Wiseman, "The Vassal-Treaties of Esarhaddon," Iraq, 20 (1 958), 1-99. Other references may be found 
in Levine. 

In addition to the Assyrian sources one should mention Herodotus and Ctesias, Xenophon in his 
Cyropaedia, as well as the Anabasis (e.g., Ill, 4, 1 1 ) and stray remarks in his Hellenica (as 1, 2, 19). Later texts, 
such as Eusebius, who gives a list of Median kings, copy the earlier sources and add little to our knowledge, 
while II Kings 17,6 and 1 8, 1 1 merely tell us that after the conquest of Samaria in Palestine by the king of 
Assyria (Sargon II), about 721 B.C., he deported the inhabitants to the 'cities' of the Medes. The Medes, 
curiously, are not mentioned in Urartian inscriptions. This is all of the literary evidence we have now, 
about the Medes, and it seems that only archaeology can provide additional information to reconstruct 
more of their history. What follows is a survey of our state of knowledge. (Other references may be found 
in the footnotes.) 



66 Chapter IV 

THE FORMATION OF THE MEDIAN EMPIRE 

The expansion of the Medes, indeed of all of the Iranian peoples in western Iran might 
be compared in analogy with the expansion of the Turks in Azerbaijan in the eleventh 
and twelfth centuries a.d. Probably, just as in the case of the Turks, so in this early 
period other Iranians migrated into the northwestern part of the plateau, together 
with the Medes. This would account for the early appearance of the Parsua in this 
area. In the Urartian inscriptions, the same land is mentioned as Barsua, devastated by 
King Menua and later by Argishti, son of Menua, from the first part of the eighth 
century B.C. 1 The location of Parsua on the southern shores of Lake Urmia had been 
generally accepted by scholars, but already in 1 962 1 had suggested rather "a district of 
imprecise location north of modern Kermanshah" as the location of Parsua 
(Akkadian: Parsuas). 2 In 1974 L. Levine reviewed the literature on the subject and 
came to the same conclusion, but he narrowed the area of Parsua to the Mahidasht, or 
the plain to the northwest of Kermanshah. 3 The question of the identity of the 
inhabitants of Parsua, however, remains. Were they the ancestors of the same Persians 
who moved south to the present province of Fars, which bears their name, or were 
they a splinter group of Persians who wandered westward from the main body 
moving south? Ghirshman in many writings supported the view that the Iranians 
came to Iran over the Caucasus route from south Russia and the Parsua of cuneiform 
inscriptions represents a stage of the march of the Persians to the south. Dyakonov, on 
the other hand, suggested that the words Parsa, Parsua and Parthia were three forms of 
an Old Iranian parsava 'rib, side, frontier,' and all three were on the frontiers of Media, 
hence, it was a common name to those peoples who settled on the frontiers of the 
central Medes. 4 Other etymologies have been proposed but none help us with the 
identification of the Parsua. Since this northern Parsua is different geographically 
from the southern Parsa (Parsu in the Akkadian versions of the OP inscriptions), one 
would normally surmise they were two groups of the same people, were it not for the 
temporal as well as geographical separation of the two. 

The first mention of the Parsua in Assyrian records is also from the time of 
Shalmaneser III (858-824 B.C.), when in his twenty-fourth year of rule he received 
tribute from twenty-seven kings of the land of Parsua. 5 Notices of this Parsua 
continue until the time of Sargon II (722-705 b.c.) when they cease. As far as I know, 
there is no reference to the northern Parsua after his reign. From the time of 
Sennacherib (705-681 B.C.) until the rise of the Achaemenids we find reference to the 
Par-su-ma-as or Parsu (both presumably for Parsua) only in the south. The name is 
mentioned together with Anshan or Anzan and other allies of the Elamites, in the 
eighth campaign of the Assyrian king against Elam. 6 Obviously, if the name 

1 G. A. Melikisvili, Uranskie klinoobrazny and Babylonia, \ (Chicago, 1926) 206 The use of 

nadpisi (Moscow, 1960), 138-40 and 218-20. For the word 'king' for these local chiefs is normal for 

a discussion of the historical implications see N. V. the ancient cuneiform records. 

Arutyunyan, Biainili (Urartu) (Erevan, 1970), 6 Luckenbill, The Annals of Sennacherib (Chi- 

134-35 cago, 1924), I, 143. A discussion of the different 

- Op. cil. [ch 3, n. 26], 47. forms of the name in cuneiform, and an etymol- 

1 L Levine in Iran, 12 (1974), 112 ogy, are found in Herzfeld, op cil 2 [ch. 3, n. 31], 

Dyakonov, Istoriya Midii (Moscow-Lenin- 727-29, but one must be cautious in using 

grad, 1956), 69. Herzfeld's works since they mix brilliance and 

5 D. D. Luckenbill, Ancient Records of Assyria nonsense. 



Medes, Scythians and Eastern Rulers 61 

disappears from the north and then reappears a few years later in the south the natural 
conclusion is that the inhabitants moved from the north to the south, but this is not 
necessarily true. We might also suppose, for example, that some, perhaps only a few, 
of the Persians in the north left to go south, or as the number of Persians in the south 
swelled, the name was transferred even though no great movement of peoples took 
place. In any case, Parsua in the north disappears from history at the end of the eighth 
century B.C., and the name is ever afterwards applied to the province today called Fars. 
For a lengthy discussion on the Parsua question see Grantovskii. 6 * Here all occurrences 
of the name are mentioned, even possible references in Classical sources to them on 
the southwest shores of the Caspian Sea, as well as Indian sources referring to them to 
the west of the Indus River. Without investigating all possible references, many of 
which are ambiguous, it would seem that members of the Parsa tribe of Iranians were 
widely scattered, whether before the Achaemenid period or during it is not possible 
to determine. The Medes, although first mentioned c. 836 B.C., later than the Parsua, 
soon became more important for the Assyrians. 

Herodotus (I, 101) mentions six tribes of the Medes, the Bousai, Paretakenoi, 
Strouxates, Arizantoi, Boudioi and Magoi. Of the six, the Paretakenoi are mentioned 
elsewhere (Polybius 31,11; Arrian 3, 1 9 ; Curtius 5,13; Strabo, 1 5, 732), and it seems 
they lived around modern Isfahan and to the south of it. An Iranian etymology could 
be reconstructed from this name, but etymologies are always hypothetical, and unless 
they give information in a larger context little can be derived from them. 7 The name 
Arizantoi has an obvious Iranian etymology 'Aryan clan' but again it tells us nothing. 
The Magi, on the other hand, present a problem since it is the same name as the later 
priestly caste, and we cannot determine whether originally they were a tribe of 
Medes who performed priestly duties, as the Levites among the Israelites, or that 
Herodotus is mistaken. Whether the Magi were centered around Rhages, modern 
Tehran, because of the religious epithets given to it in the Avesta, implying a special 
place in Zoroastrian tradition, again is pure conjecture. How much the Medes 
mingled with the local population as compared with the Persians in the south, is also 
open to speculation. In any case, we may assume that the people called the Medes had a 
number of tribes and they settled in various parts of western Iran. 

From Assyrian annals it is clear that these early Iranians on the plateau were not 
united, but lived under minor chieftains, whom the Assyrians called kings (far). We 
have already mentioned that in the reign of Shalmaneser III (c. 834 B.C.) he received 
tribute from the kings of Parsua and in his thirtieth and thirty-first years he marched 
against them and punished them. It was not long after this time that Urartu became a 
major power and rival of Assyria under King Menua (c. 810-786 B.C.) and Assyria 
was put on the defensive in the east until the revival of Assyrian power under Tiglath- 
Pileser III ( c . 745-727 B.C.). This monarch penetrated deeper onto the Iranian plateau 
than his predecessors, and he received gifts from the chiefs of the Medes as far as Mt. 
Bikni. 8 Much has been written about this mountain which has been identified as Mt. 
Demavend, but an identification with Mt. Alvand near Hamadan, or another peak of 

' E. A. Grantovskii, Ranyaya Isloriya Iranskykh kinu, etc., cf. Sumer, 7 (1956), 1-2. On the 

lemen (Moscow, 1970), 133-83. * etymologies of the names of the Median tribes see 

There were several Assyrian forms which Herzfeld, op. cit. [ch. 3, n. 31], 724-26. 

probably represented this name Partakku, Parita- 8 Luckenbill, Records [n. 5], 281. 



68 Chapter IV 

the Zagros chain, seems more likely. 9 The Assyrian contacts with the plateau were at 
first raids for booty, but then they developed into overlordships through tributaries. 
There is no evidence that in the many valleys and plains of western Iran the Assyrians 
sought direct rule through Assyrian governors and permanent garrisons. Local rulers 
instead were made Assyrian vassals. Epithets given to the Medes in Assyrian 
inscriptions such as 'mighty Medes' or 'distant Medes' do not imply any political or 
ethnic differentiation among the Medes, although it is clear that the Assyrians 
regarded the Medes as powerful adversaries even before they united to form a state. 
But the Assyrians had other problems in the east. 

An old enemy again measured strength with the Assyrians, and the annals of the 
Assyrian kings are now filled with accounts of expeditions against Urartu. Already in 
the thirteenth century B.C. Assyrian sources mentioned the Urartians as strong 
opponents of Shalmaneser I. (c. 1272-1243 B.C.). 10 First a union of tribes, in the land 
between the three lakes, modern Van, Urmia and Sevan, this area became the center 
of a powerful kingdom, the Ararat of the Bible. The various forms of this name, as 
well as other designations of Urartu, are discussed by Melikisvili 13—22, and 
Piotrovskii 33, but, contrary to several scholars, there is no connection between the 
names Urartu and the word 'Hurrian.' The early expeditions of the Assyrians against 
the Urartians in the succeeding reigns ofTiglath-Pileser I (c. 1114-1076 B.C.) Assur- 
bel-kala (c. 1073-1056 B.C.), Adad-nerari II (c. 911-890 B.C.), Tukulti-ninurta II (c. 
889-884 B.C.) are somewhat monotonous in their formulae of victories and booty 
secured. Both the early Assyrian and Urartian states had periods of alternate 
flourishing and decay in this time, but we cannot follow the histories of these peoples. 
Under Assur-nasirapal II (c. 883-859 B.C.) Assyria revived its imperial ambitions, and 
it is under his successor, Shalmaneser III, as we have noted, that both Parsua and the 
Medes first appear in Assyrian annals. At the same time Urartu experienced a revival 
and soon disputed control over lands on the Iranian plateau with Assyria. 
Unfortunately, it is not possible to determine the succession and length of reigns of 
the Urartian kings from Assyrian sources, but the names Aram and Sarduri are 
mentioned as royal opponents of Shalmaneser III. The Assyrians seem to have had the 
upper hand until the reign of the Urartian king Menua (c. 810-786 B.C.) who built 
many fortresses to defend his land against the raids of Adad-nerari III (810-783 B.C.). 
The small states and towns between the two powers, however, suffered most, for they 
were the objects of conquest and domination by both sides, and it is very difficult to 
determine the extent of either Assyrian and Urartian rule on the plateau at any time. 
Much has been written about the geography of place names on the plateau conquered 
by, or merely reached by, Assyrian armies. The mother of Adad-nerari III may be 
identified as the famous Semiramis (Shammuramat), who for a time acted as a regent 
for her son, but the legends about her in Ctesias, Moses of Chorene and elsewhere 
unfortunately cannot be used for any historical reconstructions. 1 ' Under her and her 
son, Assyrian armies raided the plateau many times, but afterwards the power of 

9 Levine. |n 3j, 1 1 9. ' ' On the ramifications of the legend see W. 

10 For the history of Urartu see G. A. Mehkis- Eilers, Semiramis, SWAW, 274 (Vienna, 1971). 
vili, Nairi-Urartu (Tbilisi, 1954); B. B. Piotrovskii, 

Vanskoe Tsarswo (Moscow, 1959); and Arutyun- 
yan, op cit. [n. 1] 



Medes, Scythians and Eastern Rulers 69 

Assyria seems to have suffered a setback for we do not hear of any other expeditions to 
the east. The next revival of Assyrian expansion to the east under Tiglath-Pileser III 
(745-727 B.C.) appears to coincide with a movement of peoples on the Iranian plateau 
and a resurgence of the Elamites, under a series of rulers such as Shutruk-Nakhunte II 
(c. 717-699 B.c). Inner conflicts, however, kept the Elamites weak, and their main 
arena of activity, as far as the sources reveal, was in Babylonia. As mentioned, in the 
north the Urartians under King Menua had begun to expand to the south and west, 
and under Argishti I, son of Menua (c. 786-764), they conquered a confederation of 
towns in the vicinity of modern Erzurum and exacted tribute. 1 2 In the inscriptions of 
the Urartian kings, copying Assyrian prototypes, booty and tribute are described in 
detail, while the burning of cities and taking captives was recorded by both Assyrians 
and Urartians. 

The early period of the expansion of Assyria onto the Iranian plateau, from c. 900- 
825 B.C. has been characterized by Levine as an attempt to control or even 
monopolize trade and trade routes to the east, while the next period to 744 B.C. saw 
the rise of a competitor, Urartu, also interested in controlling trade especially in the 
northern part of the plateau. 1 3 Certainly trade, or perhaps also the amassing of booty, 
an important source of gain in the ancient world, must have played an important role 
in the foreign policies of both Assyria and Urartu, but other factors including a quest 
for fame, may have spurred the rulers to conquest. Whereas Tiglath-Pileser III 
directed his armies along the Khurasan road through modern Kermanshah and many 
times overran Parsua and received tribute from the Medes, Sargon (722-705 B.C.), 
turned to the north. Both rulers, however, departed from the ancient custom of 
turning local rulers into vassals and instead put Assyrian governors in control of 
newly conquered areas. 14 The policy of the Assyrians seems to have shifted from 
nominal control through vassals to direct rule over the trade routes to the east. Horses 
were important items of trade but also cattle, sheep, lapis lazuli and other semi- 
precious stones, and other wares, were sought by the Assyrians. In 715 B.C. Sargon 
captured a local Median chief called Dahyuka, together with his family, and deported 
them to Hamath in Syria. This is the first mention of the name which appears in 
Herodotus as Deiokes, who presumably was intriguing with Manna or Urartu 
against the Assyrians. 15 Much has been written about the identity of the Dahyuka of 
Sargon's annals, but most scholars agree that he cannot be the founder of Median 
unity as stated by Herodotus. 16 Herodotus (1, 102) says he ruled for fifty-three years 
and was succeeded by Phraortes, who ruled for twenty-two years, followed by 
Cyaxargs, forty years, and then Astyages, thirty-five years. If one figures backward 
from the date of the rise of Cyrus, then Deiokes should have ruled c. 728-675 B.C., but 
problems arise in the chronology. Several scholars have investigated the rise of the 

1 2 Melikisvili, op cit. [n. 10], 234-35. Arutyun- ,4 Luckenbill, Records 1 [n. 5], 281 ; 2, 29. 

van, op. cit. [n. 1], 186. This is the beginning of ,5 Ibid., 2, 28. For a discussion of the name, cf. R. 

Urartian expansion. Schmitt, "Deiokes," Attzeiger der phil.-hist. Klasse 

L. Levine, "East-West Trade in the Late Iron der Osterreichischen Akademie der Wiss. 110 

Age : a view from the Zagros," Le plateau iranien et 1 973), 1 37-47. 

lAsie centrale des origines a la conquite islamique, l6 Ace. to Grantovskii, op. cit. [n. 6a] 251, the 

colloques internationaux du CNRS, no. 567 (Paris, Dahyuka captured by Sargon was ruler of Misi, a 

'')» 171-79. province of Manna. 



70 Chapter IV 

state of the Medes, but no one has reached a satisfactory conclusion ; one suggestion 
has it that Phraortes should change places with Deiokes. 17 The Cimmerian and 
Scythian invasion of northwestern Iran, however, further complicates the picture. 

CIMMERIANS AND SCYTHIANS 

The Cimmerians lived north of the Caucasus mountains in south Russia and probably 
were related to the Thracians, but they surely were a mixed group by the time they 
appeared south of the mountains, and we hear of them first in the year 714 B.C. after 
they presumably had defeated the Urartians. 18 Rusa, son of Sarduri, who ruled c. 
735-713 B.C., was the Urartian ruler who suffered at the hands of the Cimmerians, 
and then from Sargon H's famous campaign of the eighth year of his rule when the 
Assyrians marched through Parsua, the Mannai country and defeated the Urartian 
army with its allies of local rulers around lake Urmia. 19 The following march of the 
Assyrian army through Urartian territory, plundering and burning, struck a 
significant blow at the expanding Urartian kingdom, which caused a temporary 
decline in its power. It was mainly the loss of allies and territory in what is today 
Iranian Azerbaijan, however, which proved more damaging to Urartian power. The 
decline and fall of Urartu, however, was gradual. Under Sennacherib (Sinaherib), 
705-681 B.C., tribute from the far-away Medes, "whose name no one among the 
kings, my fathers, had (ever) heard" was received, and they were made tributary. 20 
This king, however, had to turn his attention from the north to the south, for the 
Babylonians and Elamites had united against the Assyrians. Elam had the support of 
Parsua (now in the south) as well as Anzan (Anshan), perhaps used as synonyms, one 
Iranian and the other Elamite, for the same area, but they lost to the Assyrians. 21 The 
Urartians under Argishti II (713-685 B.C.) then took advantage of Assyrian 
preoccupation in the south to extend their influence, if not outright conquests, to the 
Caspian Sea. 22 They never regained their earlier influence south of Lake Urmia, 
however, and the next ruler Rusa II (685-645 B.C.) turned his attention to the west in 
Asia Minor. The position of the Cimmerians in the great power struggle between 
Assyria and Urartu is not easy to determine, but it seems that they moved westward 
against Phrygia and into Cappadocia, from whence probably the name Gomer came 
into the Bible and Gamirk ' in Armenian. After the Cimmerians, however, came the 
Scythians. 

Herodotus (1, 103-04) says that the Scythians had driven the Cimmerians out of 
Europe and then followed them, but more to the east they encountered the Medes, 
whose nascent kingdom they destroyed, and then they ruled for twenty-eight years 
(106). One problem in the Assyrian sources is the term Utnmdn-manda used, 
apparently, for any barbarians in the north and east in Assyrian eyes ; so it is difficult to 

R. Labat, "KaStariti, Phraorte et les debuts de Royal Correspondence of the Assyrian Empire, 1 

l'histoire Mede," J A, 249 (1961), 3. In the same (Ann Arbor, 1930), no. 12. The form of the name 

issue of the J A E. Cavaignac writes "A propos du in Assyrian is Gi-mir-ri. 
debut de l'histoire des Medes," 153-62. 19 Cf. Arutyunyan, op. cil. [n. 1], 300 foil. 

18 Pinches in JRAS (1913), 609, and I. M. 20 Luckenbill, Annals [n. 6], 29, 68. 

Dyakonov, "Assiro-Vavilonskie Istocniki po istorii 21 Ibid., 88, 91. 

Urartu," VD1, 2 (1951), 339, also L. Waterman, 22 Arutyunyan, op. cil. [n. 1], 322. 



Medes, Scythians and Eastern Rulers 71 

tell whether Medes, Cimmerians or Scythians are meant. The movements of various 
peoples to the north and east of the Assyrians at this time are impossible to trace, but 
the entire period can be characterized as one of change and displacement of ethnic 
groups, mostly nomads moving into settled lands together with the rise and fall of 
alliances and allegiances. 

Already under Sargon we hear of a Median chief called in Akkadian Uksatar who 
sent tribute to the Assyrians. His name could be an apocopated form of *Huvaxstra 
in Iranian, or Cyaxares in Herodotus, but he was probably not related to the dynasty 
which later united the Medes, since the name must have been fairly common among 
them. 23 During the reign of Sennacherib Median chiefs, however, were undoubtedly 
extending both Median rule and Median settlements on the plateau. The Mannai 
were brought under Assyrian control by Sennacherib, but they were soon to pass 
under Median suzerainty in the time of Asarhaddon (681-669 B.C.), successor to 
Sennacherib. We can dimly discern the process of consolidation of Median power on 
the plateau, for soon old names would vanish to be replaced by Median names. In this 
process undoubtedly much of ancient traditions and practices would be preserved but 
integrated into a new Median society and culture. 

The movement of peoples to the north and east of Assyria continued under 
Asarhaddon, who was worried about them. The name Kastariti is mentioned in oracle 
texts of Asarhaddon together with Cimmerian and Mannai troops. 24 The Scythians 
also appear in the texts, called IskuzajUquza, or in the annals of Asarhaddon Aikuza, 
and a Scythian chief Ispaka, ally of the Mannai, is mentioned later in the sources. 25 In 
Herodotus (1, 103) Madyes, son of Protothygs, a Scythian king, is said to have invaded 
Media, and it is interesting that we find in an Assyrian source the name Bartatua (or 
P-) who sought the hand of the Assyrian king's daughter in marriage. 26 Although it is 
unrecorded, he was probably successful, since the Scythians seem to have become 
allies of the Assyrians for a period of time. The further history of Scythians and 
Cimmerians in Anatolia is outside the scope of this book, but in Iran we hear no more 
of the latter while Scythians, called Saka in Iranian, were probably soon absorbed by 
their kinfolk, the Medes. The presence of the Scythians on the northwestern part of 
the Iranian plateau, however, has been traced in material remains from archaeological 
excavations, especially in the hoard from Sakkiz and in Hasanlu. 27 One feature of 
material culture peculiar to the Scythians, and the Cimmerians, in this period is a 
three-lobed arrowhead, which has been found in many excavations and graves from 
south Russia to Media. This type of arrowhead has been studied by many scholars, 
the most recent summary of the evidence appearing in 1977. 28 This, together with 

23 Grantovskii, op. cit. [n. 6a], 316. literature, see A. Godard, Le trisor de Ziwiyi 

24 G. A. Knudtzon, Assyrische Cebele an den (Haarlem, 1950) esp. 125, and R. Ghirshman, The 
Sonnengott, 2 (Leipzig, 1893) no. 1. Trans, in I. M. Arts of Ancient Iran (N.Y., 1964), 98-125; R. D. 
Dyakonov, "Assiro-va vilonskie istocniki po Istorii Barnett, "The Treasure of Ziwiye," Iraq, 18(1 956), 
Urartu," VDI (1951), no. 2, 3, 4. 111-16. For Hasanlu see the bibliography in 

25 Dyakonov op. cit. [n. 24], no. 35. For a Porada, op. cit. [ch. 3, n. 2], 264. 

discussion of the names Scyth-Saka-Ashkuz and the 28 S. Cleuziou, "Les pointes de fleches 'scythi- 

miswriting Ashkenaz see I. M. Dyakonov, Istoriya ques' au proche et moyen orient," Le plateau iranien 

Mid" [n. 4], 243. et l'Asie centrale, CNRS (Paris, 1977), 187-99, 

6 Dyakonov, VDI (1951), [n. 24], no. 29. with an ample bibliography. See also B. B. Pio- 

The treasure of Sakkiz has produced a large trovskii, Vanskoe Tsarstvo (Moscow, 1959), 240. 



12 Chapter IV 

horse burials in kurgans, seems to identify the presence of nomads among the 
population, whereas the question of the originators of the notorious 'animal style' in 
bronze and other metals is more controversial. As evidence accumulates it is clear that 
the nomads had a great influence on the artistic tastes of the settled people of Iran, even 
though the actual presence of the nomads in any one area may be uncertain. It seems 
clear that Transcaucasia and Iran had much in common as far as iron weapons, horse 
trappings, horse burials, and even pottery, are concerned in the first half of the first 
millennium B.C. This is not unexpected, for obviously the rise of iron must have 
replaced bronze prototypes all over the area but hardly before the ninth - eighth 
centuries B.C. Whether nomads for the most part caused this homogeneity of culture 
in Transcaucasia and in Iran is uncertain, but they obviously contributed to it. 29 The 
question whether the Scythians, in their invasion of Iran from the north followed the 
same path as the Medes and Persians earlier, cannot be answered, but the Scythians, we 
can surmise, according to archaeological evidence, did follow the western shore of the 
Caspian Sea to the south. An earlier influx of Iranians by the same route has not been 
found. 

Herodotus (1, 106, also IV, 12) writes that the Scythians ruled over Asia for twenty- 
eight years until Cyaxares and the Medes killed (their chiefs) at a banquet and 
recovered their independence. There is no evidence elsewhere for this statement, but 
at the same time there is no reason to doubt the power and rule of the Scythians, since 
material traces of them are plentiful, while the Assyrians under Assurbanipal (668- 
632 or 626 B.C.), son and successor of Asarhaddon, did not rule the country of the 
Mannai or much else on the northern part of the plateau. By the end of the reign of 
Asarhaddon much of Assyrian authority may have been eroded, but at least many 
chieftains concluded vassal treaties with him, promising to continue to support his son 
and successor. 30 The treaties mattered little, for it was during Assurbanipal's reign 
that Assyrian power drastically declined and the Medes became the real heirs of the 
Assyrian position first on the Iranian plateau and then elsewhere. The powerful 
northern enemies of the Assyrians, however, were to suffer first from the changes. 

THE END OF URARTU 

Enough inscriptions now have been recovered to reconstruct the succession of the last 
kings of Urartu after Rusa II. Urartu was on the defensive and hardly able to maintain 
a semblance of past glory under Sarduri III, son of Rusa, who was in turn followed by 
another Sarduri IV then Erimena, whose relationship to the former kings is 
unknown. Erimena was succeeded by his son Rusa III, after whom a final ruler either 
called Rusa IV or Sarduri IV appears in texts. 31 In spite of many uncertainties about 
the succession of rulers, it seems that the kingdom of Urartu fell to the Medes 
probably in the time of Cyaxares about the turn of the seventh century B.C. The 

29 Well documented in M. H. Pogrebova, Iran i ■>' N. V. Aroutiounian, "La derniere periode de 

Zakavkaz'e v rannem zheleznom veke (Moscow, l'histoire d'Urartu," ^4/4H, 22 (1974), 428, where a 

1977), esp. 168-73. bibliography is given including the differences in 

D. J Wiseman, 'The Vassal-Treaties of views between the author and I. M. Dyakonov. 

Esarhaddon," Iraq, 20 (London, 1958), 13 (also Here we follow Aroutiounian. 
separate publication). 



Medes, Scythians and Eastern Rulers 73 

capital city Tuspa, (modern Van) was captured about 609 B.C. probably by the Medes, 
according to Dyakonov. 32 Archaeology fortunately has not only recovered Urartian 
inscriptions, but also has given us a good picture of Urartian architecture, town 
planning and defenses. The prominence of Urartian massive stonework had an 
influence on later Achaemenid as well as on Armenian architecture. 33 It was not only 
the Medes, however, who dealt a blow to Urartu in the south, but also the Scythians 
and other invaders from the north, who destroyed many towns of northern Urartu, 
including Teishebaini, present Erevan, about 590 B.C. 34 

The real heirs of the Urartians, however, were neither the Scythians nor Medes but 
the Armenians. Much has been written about the origin of the Armenians, but an 
attractive attempt to identify the Armenians, who today call themselves Hayk ' , as 
descendants of people living in the territory of Hayasa, north of Assyria, has not been 
generally accepted. 35 The Indo-European Armenian language does seem to be related 
to the Thraco-Phrygian group of languages, which implies a movement of people 
from west to east through eastern Anatolia, although the ethnic composition and 
dates of movement are unknown. Dyakonov has suggested that the word Hay- was 
used by the Urartians for westerners who had belonged to the Hittite empire 
(khattini) and the country west of the Euphrates was Khate, which the Arameans to 
the south called *Armina, which term then was later borrowed by the Persians. 36 He 
concludes that the Armenians were the result of a mixture of the inhabitants of the 
Urartian kingdom, which included Hurrians and Luvians as well as Urartians, 
together with Indo-European speaking proto-Armenians who may have been a 
mixture of people called Mushki and others who belonged to the Thraco-Phrygian 
linguistic family, plus some Scythians and others as well. 37 

We may conclude that the Armenians expanded to much the same geographical 
extent as the Urartians, and the Armenian language eventually was adopted by the 
old Urartian population. When Darius in the Old Persian version of his Behistun 
inscription mentions Armenia, the Elamite version has the same, whereas the 
Akkadian has the older Urastu, an acknowledgment of continuity and conservatism 
in Babylonia. 

Much controversy has existed over the role of Urartu in the development of 
Achaemenid culture, together with the debate on the route followed by Medes and/ 
or Persians onto the Iranian plateau. Obviously, if one or both of the great Iranian 
tribes had come over the Caucasus, the Urartian influence on the later customs, 

2 I. M. Dyakonov, "Poslednie gody Urartsogo 34 V. V. Piotrovskii, Karmir Blur (Leningrad, 

Gosudarstva," VDI, no. 2 (1951), 29-39. This is 1970), and Arutyunyan (Harut'eunean), op. cit. [n. 

also the opinion of W. Kleiss, "Zur Ausbreitung 1], 334. 

Urartus nach Osten" /slanfcu/er Mi'Kei/tm^en, 19/20 35 G. B. Djaukyan in Istoriya Armyanskogo 

(1969-70), 135. Naroda (Erevan, 1951), 25, and his "The Hayasa 

Cf. W. Kleiss, ed.,Bastam (Berlin, 1979), and Language and its Relation to the Indoeuropean 

his Topographische Karte von Urartu (Berlin, 1976). Languages," AO, 29 (1961), 353-405, esp. 398. 
The excavations of Karmir Blur (Red Hill) at 36 I. M. Dyakonov, Predistoriya Armyanskogo 

Erevan by B. B. Piotrovskii and K. L. Ogenesiyan Naroda (Erevan, 1969), 231-36. His surmises go 

u V ^ Erevan ' l 95 °-55) have produced much on farther than others and may be accepted until 

the material culture of the Urartians. See also the shown to be in error. 

results of excavations in the picture album of K. 37 Ibid., 243. On the whole, this work by 

Hovhannisean, Erebunii Orsnankarnera, The Wall- Dyakonov is the most acceptable study of the 

Paintings of Erebooni (Erevan, 1973). origins of the Armenians. 



74 Chapter IV 

institutions and culture of the Iranians would have been overwhelming. W. Kleiss, 
the expert on Urartian architecture and archaeology in Iran, has shown the Urartian 
influences on Achaemenid stone structures such as the Zendan at Pasargadae and the 
Ka 'bah of Zoroaster at Naqsh-e Rustam. 38 He described specific details of architecture 
in a convincing attribution of Urartian influence, whereas previously R. Ghirshman 
in a more impressionistic manner had claimed an Urartian origin for the cyclopean 
architecture and platforms of Masjid-e Sulaiman in Khuzistan and at Pasargadae. 39 
Given the architectural connections, and perhaps the borrowing of the title 'king of 
kings' by the Achaemenids from Urartu via the Medes, does this justify what 
Ghirshman called an Irano-Urartian 'koine' implying a fusion at least of the cultures 
of the two peoples, if not more? 40 The answer to this question lies with future 
archaeological finds to determine the path of the Iranians onto the plateau and their 
spread over it. Since the Medes were the first to form a centralized state in 
northwestern Iran, however, borrowing from their northern neighbors the 
Urartians would not be unexpected. Specific points of influence, however, are very 
difficult to establish, and one must be content at present with the probability of 
borrowings, which by no means 'prove' that the Iranians came over the Caucasus to 
the plateau. 

THE FALL OF ASSYRIA 

We have already mentioned the chronological problems which arise with the early 
Median kings. The chronology of their rule proposed by Labat, combining 
Herodotus and meagre Akkadian sources, on the whole has been accepted by most 
scholars with the first ruler after an uncertain Deiokes being Phraortes (OP 
Fravartis, Akk. Parumartis, Elamite Pirrumartis), who ruled c. 647-625 or 623 B.C. 
during the Scythian domination. 41 This ruler probably is the same as the Kastariti of 
the annals of Asarhaddon, and the difference in names can best be explained as the 
former being his personal name and the latter his 'throne' name, meaning 'possessing a 
kingdom,' Iranian *Xsa9rita. The name Kastariti, however, appears only in the omen 
texts of Asarhaddon, and this has caused some scholars, such as Dyakonov, to postulate 
an earlier rule for Phraortes from c. 675-653 B.C. followed by the Scythian 
interregnum of 653-625. 42 It is, of course, possible that Phraortes did have a very 
long rule from c. 675-625 B.C., including the Scythian domination of northwestern 
Iran, but it may be another person according to Labat, and without new sources it is 

38 W. Kleiss, "Zur Rekonstruktion des urartai- 41 Labat, op. cit. [n. 17], 3. The lengths of rule of 
schen Tempels," htanbukr Mitteilungen, 13/14 the Median kings are derived from Herodotus and 
(Berlin 1963-4), 12-14. The false windows and the absolute chronology from cuneiform records 
the crenated stone tops to the massive, square interpolated with Herodotus. On the unreliability 
temples are the chief points of resemblance. of Herodotus on the Median kings see R. R. Helm, 

39 Ghirshman, op. cit. |n. 27], 296. "Herodotus' Medikos Logos and Median History," 

40 The title 'king of kings' had existed all over Iran, 19 (1981), 85, 90. 

the ancient Near East, but in the seventh century 42 1. M. Dyakonov, Istoriya Midii [n. 4], 266, 

B.C. it seems to have been in vogue only in Urartu, 282. 
according to written records; cf. Frye, The 
Charisma of Kingship in Ancient Iran," IA, 6 
(1964), 111-15. 



Medes, Scythians and Eastern Rulers 75 

not possible to reconstruct a satisfactory chronology of the early Median kings. We 
can only surmise that the Median state came into formation during the time of 
Phraortes in the middle of the seventh century B.C. although Dyakonov would date 
the declaration of independence of Phraortes from Assyrian rule to the year 673 B.C. 43 

With the next ruler we are on firmer ground, for Phraortes may have been killed 
by the Assyrians according to Cavaignac, but in any case he was succeeded by 
Cyaxargs (OP *Huvaxstra, Akk. Umakistar). 44 Herodotus (1, 106) said that he ruled 
forty years, but this was most likely after the Scythian domination, and not before it as 
the historian thought. According to Herodotus (1, 103) Cyaxares was the first ruler to 
divide an army according to weapons used, separating lance bearers from bowmen 
and the like. This information cannot be correct as stated, and one should ask what the 
historical reality behind it is. As Dyakonov persuasively argued, it is probably the 
change from a force of tribal levies to a regular army which led to the remark of 
Herodotus. 45 Assyrian sources, unfortunately, do not help in elucidating the rise of 
the Median state, for during the reign of Assurbanipal Assyria fought many times 
against Egypt, Babylonia, Arab tribes, and Elam, not giving much attention to the 
Iranian plateau. Elam was destroyed as an independent power under Assurbanipal, 
and so too was the kingdom of the Mannai. 46 One may conjecture that the 
destruction of both Elam and the kingdom of the Mannai by the Assyrians, together 
with the fall of Urartu gave the Iranians a chance to assert their power, which was 
soon felt by their neighbors after the death of Assurbanipal. 

A strong ruler, Nabopolassar, became king of Babylon about 626 B.C., so Assyria 
was in danger from the south as well as from the Medes to the east. Unfortunately, we 
have no information, literary or archaeological, about events on the Iranian plateau in 
this period and cuneiform records also tell us little about Mesopotamia, so we are 
reduced to conjecture. 47 Just as we may assume that with the destruction of the power 
of Elam, Persian leaders took their place in Fars, if not also elsewhere, so we may 
conclude that the Medes consolidated their state at the expense of the Mannai, and of 
course of various Iranian tribes. It is quite possible that Cyaxargs subjected the Persians 
in the south to Median rule in the period around 625 B.C. as Herodotus (I, 102) 
suggests. We must stress again that all is conjecture and must be fitted into a time scale 
which does not clash with earlier and later data. 48 Luckily for the years just before and 
after the fall of Nineveh we do have important cuneiform sources in the British 
Museum which give a summary of principal events relating to Assyria in the period 

Ibid., 266. Such precise dating on the basis of ment of the army of the Achaemenids was ascribed 

oracle texts, however, is not convincing. to Medus, presumably a Median king in the mind 

E. Cavaignac, "Sur un passage de la tablette of the poet. 

B.M.25 127," Revue d'Assyriologie, 51 (1957), 28- 46 Luckenbill, Records, 2 [n. 5], 328-29. The 

"• The tablet is full of lacunae, and relevant to the Mannai were made subject to the Assyrians and 

Medes it only has, "he turned his face towards thus lost their independence. 

Nineveh," followed by uncertain accounts of 47 Apparently Nabopolassar began hostilities 

skirmishes. Nowhere is a proper name mentioned, against Assyria shortly after his accession and there 

and Cavaignac assumes that 'he' refers to the was constant fighting; cf. S. Smith, Babylonian 

Assyrian king, and the skirmishes refer to Phraor- Historical Texts (London 1924), 23. 

«s, following Herodotus. 48 D. Stronach, "Excavations at Tepe Nosh-i JJn 

45 Dyakonov, Istoriya Midii [n. 4], 295. In 1967," Iran, 7 (1969), 6, reviews several opinions 

Aeschylus, The Persians, verse 765, the establish- of events during this period. 



16 Chapter IV 

of its fall. 49 It is not possible here to give an account of the rise of Babylonian military 
might and the decline of the Assyrian forces, but the steady advance of the 
Babylonians year by year against Assyria is well chronicled. In 61 5 B.C. the old capital 
of Assyria, the city of Aisur, was besieged but not taken by the Babylonians, and in 
October of the same year the Medes raided the province of Arraphu (modern 
Kirkuk). 50 The following year the Medes after marching to Nineveh turned south 
and captured Assur which they plundered. Then, in the translation of Wiseman, 
"[The king of Ak]kad and Cy[axar]5s met one another by the city. They established 
(an alliance) of mutual friendship and peace. [Cyaxarjes and his army returned to his 
land." 51 Although the text is poorly preserved for the events of the year 612, it is clear 
that in the summer the capital Nineveh fell to the Babylonians and Medes and the 
Assyrian king was killed, while his city was plundered, after which the Medes 
departed. Another Assyrian king proclaimed his rule in the town of Harran and with 
Egyptian aid maintained his independence until the summer of 610, when the 
Babylonian and Median allies captured the city. Skirmishes continued between the 
Egyptians and Babylonians until the Egyptians were defeated at a battle of 
Carchemish in 605 and peace was established for a time; thus the long domination of 
Assyria in the Near East came to an end. 

REMAINS OF THE MEDES 

Very little has survived which can be confidently attributed to the Medes, and 
excavations of Median sites are few. Until the present site of Hamadan, ancient 
Ecbatana, is excavated we must rely on the small fortress-site of NOsh-e Jan, and a 
series of rock-cut tombs or other sites with possible Median layers or influences such 
as Godin Tepe and Baba Jan Tepe. Fortunately the 'Median' rock-cut tombs have been 
discussed in detail and only a few remarks may be of interest here. 52 The five 
principal tombs which have been dated to Median times, and are said to be tombs of 
Median princes, are Kizkapan in Iraqi Kurdistan, Ferhad o SirTn at the village of Sahna 
near Kangavar, Fahraqah (or FakhrTqah some 10 km. northeast of Mahabad) near the 
village of Andarqash, Dukkan-e DaOd, 3 km. south of Sar-e Pul 2ohab, and Sakawand 
or Ishaqvand, 3 km. north of Deh-e No, which village is 18 km. southwest of 
HarsTn. 53 Because of the poor stone carving on all of these, relative to the impressive 
Achaemenid tombs at Persepolis and Naqsh-e Rustam, it was generally thought that 
the above tombs were Median in date. Von Gall, however, convincingly argued that 
all of them are rather attempts to copy royal Achaemenid tombs and are to be dated to 
the late Achaemenid period, although they may well be the tombs of local Median 

49 Cf. D. J. Wiseman, Chronicles of Chaldaean storming of Nineveh to Manda tribes led by 
Kings 626-556 B.C. (London, 1956). Arbakes rather than to Medes, is unacceptable. 

50 Ibid., 13 and 57. The Mannai fought for the 52 H. von Gall, "Zu den 'Medischen' Felsgra'bern 
Assyrians. On the Mannai prisoners of the in Nordwestiran und Iraqi Kurdistan," ArchHolo- 
Babylonians see p. 55. Dyakonov, Istoriya Midii [n. gischer Anzeiger, 81 (1966), 19 foil. 

4], 304, thinks this defeat of Mannai forces gave 53 I give official and local names and geographi- 

Cyaxares the opportunity of incorporating their cal locations of several of the sites since many who 

land in his domains. have written about them are either vague or 

51 Wiseman, op. cit. [n. 49], 59. The attempt of erroneous in these matters. 
K6nig in op. cit. [ch. 1, n. 34], 42, to ascribe the 



Medes, Scythians and Eastern Rulers 77 

chiefs. 54 With the tombs unable to serve as evidence for Median work, we turn to the 
Zendan-e Sulaiman, the remarkable crater at Takht-e Sulaiman in Azerbaijan, which 
was undoubtedly a holy site, for around the conical mountain, which was filled with 
water at this early period, houses, or rooms for piligrims or priests had been built. The 
high terrace on one side seems to confirm Herodotus (1, 131) when he said that the 
Persians do not erect statues to the gods, temples or altars, and they offer worship to 
Zeus and other deities on the tops of mountains. The pottery found here, however, 
mostly dates from the eighth century B.C., and the excavator believes the Zendan was 
a Mannai site which was destroyed by the Medes. 55 After a survey of a small site 
northeast of Maku near a village called Pul-e Dast, Kleiss remarked that "der Bau in 
Corbulaq[ist] bis heute das einzige Gebaude im Nordteil von Iranisch-Azerbaidjan, 
das von den Medern errichtet worden ist." 56 In a building there he was able to find an 
early prototype of the court-ayvan plan of later Iranian buildings. When we turn to 
the artistic finds from Hasanlu and Ziwiye, it seems that again they are remains of the 
Mannai, with possible Scythian (or Median?) influences on the style. 57 In the art it is 
clear that the two primary strains are influences from the ancient Near East, from 
Assyria in particular, and the nomadic or 'Scythian' style. What we are unable to 
discover is the relation of Median art to this "Scythian' style, and one may tentatively 
conjecture that the Medes, in their early spreading over central and northwestern 
han, hardly could be distinguished from Scythians by the local, settled population. 
The Median tribes undoubtedly absorbed much from the settled cultures of the 
Mannai, Urartians, Caspians, and other peoples, and in culture as well as political 
supremacy, they were the heirs of these peoples as the Persians were of the Elamites in 
the south. Unfortunately we know very little about the composition of the 
population on the Iranian plateau at the time of the fall of Nineveh. Presumably the 
population had been mixed to a great extent, for one reason by the policy of the 
Assyrian kings to deport rebellious peoples and settle them in widely scattered parts of 
the Assyrian domains. From the Bible (Kings II, 17, 6 and 18, 1 1) we have noted that 
Sargon II, and presumably Sennacherib, both deported Israelites to the 'cities' of the 
Medes. Others must have felt the wrath of the Assyrians in a similar manner, and the 
brisk trade quite likely also brought about a small movement of peoples. The spread 
of qanat irrigation technology must have opened new areas to cultivation on the 
plateau, and one may assume an increase in the population. Dyakonov is probably 
correct in denying the existence in Media in this period of great latifundia where 
large numbers of slaves were needed, and although the Medes may have brought 
captives home from the Assyrian wars, there is no evidence of any large-scale 
displacement of peoples. 58 That various peoples maintained their identities after the 
fall of Nineveh is suggested by the book of Jeremiah 51, 27-28, where the prophet 
speaks of the kingdoms of Ararat (Urartu), Minni (Mannai) and Ashkenaz 

54 von Gall, op. cit. [n. 52], 38. R. M. Boehmer, "Volkstum und Stadte der 

55 W. Kleiss, Zendan-i Sulaiman, die Bauwerke Mannaer," Mitleilungen des Deutschen Archaotogi- 
(Wiesbaden, 1971), 68. Urartian temples and schen Instiluts, Abteilung Baghdad, 3 (Berlin, 
others were also placed on heights. 1964), 11 foil. 

56 Kleiss, "Zwei Platze des 6. Jh.s v. Chr. in 58 Dyakonov, htoriya Midii [n. 4], 312. The 
Azerbaidjan," AMI, 9 (1976), 116. Medes did not follow the Assyrians in this regard. 



57 



Porada, op. cit. [ch. 3, n. 2], 108-36. See also 



78 Chapter IV 

(Scythians) with the kings of the Medes, opposing Babylon. Presumably this passage 
refers to the time after 597 B.C. when the Babylonians took Jerusalem. If any credence 
is to be given this statement, we may suppose that the Median state was really more of 
a confederacy of various rulers than a centralized monarchy. 

From the excavations at Nush-e Jan Tepe, and from Assyrian representations of 
Median towns on reliefs, it is clear that settlements were highly fortified, a necessary 
precaution during the long period of Assyrian raids on the plateau and tribal warfare. 
Archaeology suggests that fortified areas were abandoned or allowed to decay after 
the fall of Assyria. 59 For any picture of Median state or society in this period, 
however, we are obliged to interpolate from later sources, such as the inscriptions of 
the Achaemenid kings, and read back into Median times certain inferences. For 
example, the Greeks borrowed the word 'paradise,' or royal park, from the Medes and 
not from the Persians, since the form of the last word is with a -z, *daiza instead of 
Persian *daida (in Ehmite /par-te-taS). This implies that the Medes had created the 
enclosed royal hunting park before the rise of the Achaemenids. Probably the group 
of Iranian tribal warriors had given place to a professional army, or at least a royal 
guard. Dyakonov would see in the Behistun inscription of Darius (II, 16), speaking of 
the Median army kara Mada haya vidapatiy aha, "the Median people which was in its 
domain(s)," a reference to the landed nobility on its lands, while the kara haya upa 
mam aha "the people which remained with me (Darius)" was the professional army 
and court nobility. 60 Whether this presupposes a division of the Median nobility 
under Cyaxargs, or his successor, into landed nobility and court nobility is 
questionable. It is perhaps better to admit the lack of information about the Medes 
and reserve discussion of later sources for later history. 

EXPANSION OF THE EMPIRE 

Herodotus (1, 134) said that the Medes ruled over their neighbors, who in turn ruled 
over those farther away, which has been interpreted as meaning that the subject 
peoples ruled themselves, sending tribute to the Medes, whereas under the 
Achaemenids the system of satraps (usually Persians) was introduced. 61 As Dyakonov 
aptly observed, the references to 'kings' under the Median king in both Jeremiah (25, 
25 and 51,11 and 28) and in a cuneiform text, implies a system of vassal rulers rather 
than a centralized state. 62 Since the OP title xlayaBiya xHayadiyanam 'king of kings' is 
probably Median in origin, though the Urartians and others used the title long before 
the Medes, obviously the king of the Medes was the overlord of other rulers. 63 What 
areas did the Medes rule? To the east we have no sources to answer this question but 

59 See D. Stronach, "Excavations at Tepe Ndsh-i 62 Dyakonov, op. cit. [n. 4], 336. 

J5n," Iran, 16 (1978), 9-10. On Bab5-j5n Tepe see " Cf. R. Schmitt, "KSnigtum im Alten Iran," 

the same issue, 41^12. Saeculum, 28 (1977), 386 and for the etymology, 

60 Dyakonov, Istoriya Midii [n. 4], 332-34. It is O. Szemerenyi, "Iranica V," AI, 5, Monumentum 
also dubious to assign the OP word spada 'army' to H. S. Nyberg, 2 (1975), 313-23. Szemerenyi is 
the Medes and kara 'people, horde,' to the Persians, wrong, however, in asserting that the title 'king of 
cf. Handbuch, 144, tahma for reference to kings' was used by the late Assyrian kings. It is 
etymologies. really found in the titulary of the early Assyrian 

61 W. W. How and J. Wells, A Commentary on kings, not those against whom the Medes fought. 
Herodotus, 2 (Oxford, 1964), 115. 



Medes, Scythians and Eastern Rulers 79 

only surmises. The tribes of the Medes already have been mentioned, and the 
Paretakeni were those whose center was present Isfahan, apparently extending into 
the desert on the east of Tabbas (Curtius V, 13, 2). Whether the Orthokorybanti, 
mentioned by Herodotus as one of the contingents in the army of the tenth satrapy of 
the Achaemenids (III, 92), were the 'high hat' Scythians who maintained a separate 
identity in Media, is unknown, but a reasonable surmise. 64 Did they have a king who 
gave fealty to Cyaxares or did they merely send tribute as more distant peoples? We 
can only conjecture that the tribes of the Medes were directly under Cyaxares 
whereas others were more independent, such as the Persians. 

The Persians, we are told by Herodotus, the Old Persian inscriptions, and a passage 
from a prism of ASurbanipal, had kings going back to an eponymous ancestor 
Achaemenes (OP Haxamanis). 65 Whether there were two kings in Fars is unknown 
but not improbable in light of the genealogy of the early Achaemenid kings 
following two lines after Teispes (OP CispiS). 66 One may further speculate that the 
line of Cyrus had its center in Pasargadae, his later capital, while his brother 
Ariaramnes (OP Ariyaramna) had his center to the southwest. We do not know 
when these local kings submitted to Median overlordship, before or after the fall of 
Nineveh, but probably under the rule of Cyaxares. Other Iranian tribes to the east are 
more problematical. 

In Kerman lived a Persian tribe, the Germanioi or Kermanioi (Herodotus 1, 125), 
which name some scholars have connected with the word 'German', while others 
have proposed various etymologies for it. 67 Presumably when Fars submitted to the 
Medes so did Kerman. In Seistan, ancient Zranka or Drangiana, one may guess a 
Median control, but again there is no evidence, and the same is true for various tribes 
which lived in the southeastern part of the Iranian plateau and which cannot be 
identified, such as the Outioi, whom Herzfeld confidently identifies with OP 
YautiyS, NPjut or jut and Arabic Zutt 'gypsies.' 68 Whoever the inhabitants of the 
desolate stretches of the southeastern Iranian plateau were, any central control from 
western Iran must have been weak, as it has been throughout history. 

Much has been made of a passage in Herodotus (III, 1 17), speaking of the area of the 
present lower Harl Rud which has been identified as the Akes River of the 
historian. 69 The Khwarazmians ruled this plain which was on the borders of the lands 

64 Dyakonov, op. cit. [n. 4], 338. His etymology, 66 V. I. Abaev, "Iz Iranskoi Onomastiki," 
meaning 'fairy' or supernatural beings is less Istoriya Iranskogo Gosudarstva i Kultury, ed. by 
likely than 'mountaineers,' proposed by Eilers, B. G. Gafurov (Moscow, 1971), 269, supports the 
"Demawend,"[ch.l,n. 19], 348, since the name is reading Cispis, while W. Hinz, Neue Wege im 
also applied to people in Central Asia. Ahpersischen (Wiesbaden, 1973), 25, upholds the 

65 In the fragment of prism from Babylon of the reading CaispiS. More weight accrues to Abaev in 
year 639 B.C., line 7 says that Cyrus, king of (the this controversy. 

land) Parsumas, heard of the great victory of 67 Cf. C. Bartholomae, "Iranisches," ZU, 4 

Assurbanipal over Elam and sent his eldest son (1926), 185, who thinks of a foreign word, 

Arukku to Nineveh with tribute. This Cyrus whereas Eilers, "Demawend" [ch. 1, n. 19], 184, 

should be the grandfather of Cyrus II, the founder gives a bibliography of discussions of the name 

of the Achaemenid dynasty. Cf. E. F. Weidner, which he considers Iranian. 

"Die Slteste Nachricht liber das persische Konigs- 68 Herzfeld, Zoroaster, 2 [ch. 3, n. 31], 733. 

haus," Archivfiir Orientforschung, 7 (1931), 4 and 69 W. B. Henning, Zoroaster, Politician or Witch- 

6. Dyakonov, op. cit. [n. 4], 349, explains the name Doctor? (Oxford, 1951), 42, where reference is 

Arukku as Iranian *Ar(ya)uka. given to Markwart. 



ov Chapter IV 

of the Khwarazmians, the Hyrcanians, the Parthians, the Sarangians (in Seistan) and 
the Thamanians (in Arachosia?). From this sentence W. B. Henning concluded, "We 
can thus be fairly certain that there was a state in eastern Iran which centered around 
Marv and Herat and coexisted with the Median Empire; which was led by the 
Khwarazmians and abolished by Cyrus, who deprived them of their southern 
provinces, whereupon they gradually retired to their northern possessions along the 
River Oxus." 70 He further puts Vishtaspa, the patron of Zoroaster, as the last ruler of 
this state, thus giving the prophet a fixed time and place. This implies no Median 
overlordship in this part of the east, a position opposed by Dyakonov and others. 7 * 
These are interesting surmises, but they remain no more, and only new found 
inscriptions can resolve the question. The likelihood of some kind of Median 
hegemony exists, but how far it extended or its nature is unknown. Just as with the 
eastern Iranians, so in the Caspian area, the extent of Median rule is impossible to 
determine. Since Median armies went far into Anatolia and, as we have mentioned, 
probably put an end to Urartu, then most likely areas of easier access, such as 
Khurasan, submitted earlier to Median rule. Only by interpolating between earlier 
Assyrian and later Persian rule can we come to some appreciation of the Medes where 
we have virtually no sources. For example, as Dyakonov points out, Achaemenid 
satrapies were both larger and more independent than Assyrian provinces, so he 
proposes that the Median satrapies (for the form of the word 'satrap' is Median and not 
Persian) were between the two in size and independence. 72 Such impressionistic 
views of the Medes, unfortunately, at present are all we have in reconstructing the 
past of the Medes. 

ART AND MATERIAL CULTURE 

From the domain of art and archaeology we gain some impressions of the Medes and, 
as mentioned, their dependence on the culture of the Urartians, Elamites, and others 
of the ancient Near East is great. From representations on Assyrian reliefs some 
scholars have concluded that early Persian soldiers wore feather headdresses, or that 
Medes invariably wore tall, round felt hats and trousers. Again, unluckily, there are 
no identifying inscriptions so the designations are speculative. If we could be sure that 
certain figures were Medes, then something could be said about the costumes, but 
since styles change, it is not easy to assign decorative features to the Medes and to no 
one else. There is little point in repeating such statements as, "Just as the Medes 
probably handed down to the Persians the elements of Scythian art which they had 
absorbed or obtained independently through eastern connections, so they must also 
have been the middlemen for the continuation in Achaemenid art of other stylistic 
traditions which prevailed in Iran in Median times;" 73 or, "we know next to nothing 
about Median architecture." 74 It is clear that without extensive excavations in 
Hamadan, on the site of the Median capital of Ecbatana, there is little hope of progress 
in our knowledge of the material culture of the Medes. 



70 Ibid., 42-43. Dyakonov, Istoriya Midii [n. 4 J, 403-12, with the 

71 Istoriya Midii [n. 4], 357-58. suggestion that the 'animal style' in bronzes was 
Ibid., 361 . Median by origin. 

"Porada, op. cil. [ch. 3, n. 2], 140. Cf. 74 Ghirshman, op. tit. [n. 27], 87. 



Medes, Scythians and Eastern Rulers 81 

RELIGION 

The Zend5n-e Sulaiman, probably a site of the Mannai, has been mentioned, and 
presumably a fire altar has been found in a room with high walls in the fortified site of 
Nush-e Jan, but this is all the material evidence we have for the religion of the Medes. 
A fire cult was no monopoly of the Medes, so the presence of an altar in excavations 
does little in explaining the uses and beliefs connected with it; even the Hittites had a 
fire cult. 75 Since the discovery of a fire altar in an excavation gives few details about 
the religion practiced at the site, we turn as usual to Herodotus, who says (1, 131) that 
the Persians (and presumably also the Medes) have no images, temples or altars (sic) ; 
they worship in the open air on mountain tops and kill their sacrifices in a place free 
from pollution, but one of the Magi must be present and chant a hymn. Further (I, 
140), the Magi expose the bodies of the dead to birds of prey, and they kill crawling 
creatures, as snakes and scorpions. Much has been written about this information of 
Herodotus, and the question of the religion or religions of the various Iranian peoples 
is always bound to a comparison of Classical sources with the Avesta and more recent 
Zoroastrian practices. Herodotus, of course, lived long after the fall of the Median 
kingdom, so, as usual, one must interpolate backwards. The three forms of disposal of 
the dead, cremation, exposure or burial have been found in excavations in various 
parts of the Iranian world, although cremation, unlike among Hindus and Vikings, is 
rare in Central Asia and unattested on the plateau. 76 Since exposure of the dead body 
has been a feature of the Zoroastrian religion for almost two millennia, there is no 
reason to reject Herodotus' assertion that this is a practice especially of the Magi. The 
desire not to pollute the earth with decaying matter seems to be the principal reason 
why a priesthood would enjoin their followers to adopt exposure in place of former 
burial in the earth. Since in a passage of Xanthos (FHG 1, 42 and Frg. Hist. HI c. 2), the 
prohibition against cremation is attributed to Zoroaster by the Persians, one may 
accept this as an early Zoroastrian practice, however, frequently breached. No matter 
when the Magi adopted the reforms of the prophet, they were priests from early 
times, perhaps a tribe of the Medes as Herodotus says, or a priestly caste just among the 
Medes and then among other Iranians. In any case, we can safely assume that the Magi 
held a prominent place among the Medes as interpreters of dreams and auguries, as 
well as fulfilling their priestly functions, for the word 'magic' comes from their 
name. 77 The process by which the Magi became the Zoroastrian priests is much 
disputed and the early periods of their activity remain dark. 78 

THE LYDIAN WAR AND FALL OF THE MEDES 

The marriage of a Median princess to Nebuchadnezzar, son of the Babylonian king, 
before the fall of Nineveh was part of the joint strategy of the two allies against the 

75 V. V. Ivanov, "Kult Ognya u Khettov," (Moscow, 1963) 127-33, and the following article 
Drevnii Mir, Festschrift for V. V. Struve, ed. by N. by G. P. Snesarov, 134-40. 

Pigulevskaya (Moscow, 1962), 271-72. Fire is 77 For an etymology of magus as 'powerful, etc' 

used in Hindu rituals and elsewhere. see J. Pokorny, Indogermanisches Etymologisches 

76 For the discovery of funeral urns in ancient Worterbuch, 1 (Bern, 1959), 695. This is not 
Central Asia see Y. A. Rapoport, "Some Aspects of accepted by E. Benveniste, Les Mages dans I'ancien 
the Evolution of Zoroastrian Funeral Rites," Trudy !"»> (Paris, 1938), 20. 

25 Mezhdunarodnogo Kongressa Vostokovedov, 3 78 Cf. Boyce, op. cit. 1 [ch. 3, n. 21], 10-11. 



82 Chapter IV 

common Assyrian foe. 79 Afterwards, however, one cannot speak of an alliance of the 
two for tensions developed over the boundaries and spheres of influence. While the 
Babylonians were conquering Syria and Palestine, driving out Egyptians, the Medes 
extended their rule in eastern Anatolia, where they came into conflict with the Lydian 
kingdom with its capital at Sardis. We do not know the circumstances leading to the 
war between the Lydians and the Medes which Herodotus (I, 74) tells us lasted five 
years until the famous battle of the eclipse of the sun on 28 May 585 B.C. after which 
peace was made through intermediaries, the ruler of Cilicia, called Syennesis, and the 
Babylonians. The Halys River or present Kizil Irmak formed the boundary between 
the two states and Aryenis daughter of Alyattes of Lydia (ruled c. 617-561 B.C.) was 
married to Astyages, son of Cyaxares, as a sign of alliance. 80 The Babylonians, 
however, cannot have regarded the growing power of the Medes with anything but 
dismay and led to the construction of a defensive wall north of Babylon, and 
predictions on the fall of Babylon by Isaiah and Jeremiah. 

Herodotus (1, 106) assigns forty years to the rule of Cyaxares while, Eusebius (Frg. 
Hist. IIB, Nr. 250) gives thirty-eight, but such numbers, in these and other ancient 
sources, cannot be substantiated and can only be reported as fitting the general 
framework of chronology. He was followed by his son Astyages, about whom we 
find much folklore and fable, relating primarily to the rise of Cyrus. 8 ' We have 
conflicting information about the family of Astyages in various sources. Herodotus (I, 
107 foil.) says that Astyages had a daughter called Mandane about whom he had a 
dream that she urinated and flooded all of Asia. Astyages became fearful and instead of 
giving her in marriage to a Mede, gave her to Cambyses of Persis, from which union 
Cyrus was born. But Astyages later had another dream, that from the womb of his 
daughter a vine grew which overshadowed Asia. Then he ordered one of his nobles 
Harpagos to kill the boy, but Harpagos instead gave him to a shepherd called 
Mitradates and his wife Spako, which is the Median name for dog. 82 The woman 
substituted her dead child for Cyrus and raised the baby as her own. Years later the 
royal blood of Cyrus became manifest and Astyages found out about him, then 
received him into his household, but he punished Harpagos by killing his son and 
serving him to his father at a banquet. Then the king sent Cyrus to Persis to his real 
father and mother who received him and, after hearing his story, let it be known 
generally that he had been saved by a dog. To shorten the story, Harpagos, bent on 
revenge, later persuaded Cyrus to revolt against Astyages, and during a battle he led 
the Medes to desert and support Cyrus which gave the victory to the Persians. 



79 See Wiseman, op. cil. [n. 49], 15. for -w-. The account of Ctesias as found in 

80 The Lydian character of the names of the Diodorus Siculus (II, 32, 6) is fantasy. 

king and his daughter are asserted by L. Zgusta, 8 - The name Mitradates is given as Atradates by 

Kleinasialische Personemiamen (Prague, 1964), 55, Ctesias in Nicholas of Damascus (Frg. Hist. II, teil 

103 but no etymologies are given. The war A, 361), and in Strabo (729) the name of Cyrus 

between the two may have been started by Median himself is given as Agradates, probably a mistake 

claims to rule over the Scythians in Cappadocia. for the same name as that found in Ctesias. Cyrus 

1 On the etymology of his name *fiti-vaiga was a noble and not the son of a shepherd. The 

"brandishing a lance," see Herzfeld, op. cil. [ch. 3, n word for 'dog' is correct and much has been 

31], 1, 90, and E. A. Grantovskii, op. cit [n. 6a], written about it. 
330. In Akkadian he is called Istumegu with -m- 



Medes, Scythians and Eastern Rulers 83 

This is the Cyrus saga, with parts of the myths of the Indo-European peoples 
integrated into it. It is not necessary to discuss the variants of the myth by Ctesias and 
the account of Justin's epitome of Pompeius Trogus, which follows Herodotus, or 
other more apocopated parts of the story of the rise of Cyrus, since the saga has been 
discussed in much detail elsewhere. 83 Binder has convincingly indicated that the 
various elements in the saga such as the abandonment of a royal child and its 
nourishment by an animal (dog, wolf, eagle or cow) are very widespread motifs 
relevant to the founder of a new dynasty, and probably bound to ancient rites of 
nomadic young warrior fraternities (Mannerbiinde) on the steppes of Central Asia. 84 
To go further, however, and draw cosmogonic and far-reaching conclusions on the 
religious significance of the myths is hazardous, even though one may accept the 
premise that the stories are much more than just tales or literature but do have cultic 
and ritualistic origins. The saga of Cyrus must have been attached to his birth and 
ancestory soon after his death, for the life of Cyrus became an almost mystic symbol 
for later rulers of Iran. The testimony of Plutarch (Life of Artaxerxes, 3) that a new 
Persian king had to be initiated into his new position at Pasargadae by following a 
ritual of wearing Cyrus' robe and eating the simple food which Cyrus ate, indicates 
the great respect for tradition of the Iranians and the value of the saga as more than a 
tale. Since a later writer, Aelian (hist. anim. 12, 21) says that Achaemenes, ancestor of 
Cyrus, as a child was nourished by an eagle, we can see the extension of the variants of 
the myth backwards in time, as well as forward to the Arsacid and Sasanian founder 
legends. The Shahname, of course, has many variants of the story of the abandonment 
of a child and its rearing by an animal. I have called the elements of the motif of the 
founder of a new dynasty in Iran part of the charisma of Persian royalty which has 
the following components: royal blood in descent from the previous dynasty, 
upbringing with common folk undergoing hardships, recognition of the charismatic 
'royal fortune,' called xvarsnah in Avestan or later farn or fan, in the youth, and his 
overthrow of the ruling tyrant to usher in a new age. To use the saga for a 
reconstruction of history, however, is an uncertain and difficult undertaking, even 
though it obviously contains much of value. The sifting of fact from fiction in the 
saga requires external sources, which unfortunately give us little more than a sequence 
of a few events. 85 So even when source material is available, it is often so mixed with 
fable and fantasy as to be suspect. 

The terse Akkadian account of the defeat of Astyages goes as follows, "[Sixth year 
• . . King Ishtumegu] called up his troops and marched against Cyrus, king of Anshan, 
in order to me[et him in battle]. The army of Ishtumegu revolted against him and in 
fetters they delivered him] to Cyrus. Cyrus (marched) against the country 
Agamtanu; the royal residence (he seized); silver, gold, (other) valuables . . . of the 

83 Cf. A. Bauer, Die Kyros-Sage und Verwandtes Si Binder, op. cil. [n. 83], esp. 29, 45, 58. See also 

(Vienna, 1882), 86 pp.; G. HUsing, "Beitrage zur R. Frye, The Charisma of Kingship in Ancient 

Kyros-Sage" OLZ, 1903-06 (reprinted as a book- Iran," [n. 40], 40-46. 

let, Berlin, 1 906), 1 59 pp. ; and finally with ample 85 The Greek didactic treatise by Xenophon, the 

bibliography, G. Binder, Die AusseUung der Cyrcpaedia, is likewise difficult to use as a source of 

Kdnigskinder Kyros und Romulus, Beitrage zur history, although he, like Herodotus, says that 

klassischen Philologie, 10 (Meisenheim, 1964), esp many stories circulated about the life of Cyrus. 
17-28, 175-82. 



84 Chapter IV 

country Agamtanu he took as booty and brought (them) to Anshan." 86 Another 
Akkadian source simply says that Cyrus "made the Guti country and all the Manda- 
hordes bow in submission to his feet," while a text describing Nabonidus' desire to 
repair the temple of Sin the Moon deity at Harran mentions how Cyrus, king of 
Anshan, vassal of the king of the Manda-hordes, with a small army will overthrow 
the far-flung Medes and take Astyaggs to his land as a prisoner. 87 From these sources 
we may conclude that Cyrus defeated Astyaggs in battle, and the Median army 
revolted against Astyaggs delivering him to Cyrus who plundered Ecbatana and took 
Astyaggs as a prisoner to Persis. If this battle took place in the sixth year of the reign of 
Nabonidus, as one text says, then we may date it 550-549 B.C. It is much better to 
follow contemporary Akkadian sources, sparse though they may be, than the 
embellished tales of Herodotus, Ctesias and their successors, who collected their stories 
from various sources, including tellers of tales. Their general accounts, however, do 
not contradict the cuneiform sources and this much we accept. 

The theory that the royal house of Media was related to Zoroaster's family has been 
proposed and elaborated by several scholars. 88 The reason for this assertion is found in 
Ctesias, who gave the name of the first husband of the daughter of Astyaggs as 
Spitamas, and later the name of the magus whom Darius killed, not Gaumata or 
Bardiya, but Sphendadatgs, both of which names belong in the family of Zoroaster. 
Herzfeld (p. 51) concludes that Zoroaster is to be identified as Spitakgs, son of 
Spitamas and Amytis, daughter of Astyaggs. If the Median royal houses were in any 
way connected with Zoroaster, however, we would have heard of this directly in the 
sources and not in certain names which modern scholars use to support their theories 
and reject others as fantasy. Ancient authors have suffered from modern interpreters 
who praise and use them when it fits their fancies, and then condemn and reject those 
same authors when their information conflicts with modern hypotheses. The name 
Spitama does appear in various sources, including probably a cuneiform tablet 
(Herzfeld, 48), but this in no way allows us to infer the connection of Zoroaster with 
all or any of the names, if they are historical. Furthermore, whereas no one should 
ignore the long continuity of families as well as traditions in Iranian history, to base 
far-reaching theories on the basis of names or etymologies alone is fraught with 
misconceptions. It is quite possible that Astyages had no son to succeed him, but it is 
also possible that the ancient Iranian kingly saga demanded a blood link between 
Cyrus and the daughter of the last king of the Medes. 89 

Although the next empire of the Achaemenids came to be called one of the Medes 
and the Persians, at the beginning of the rule of the Achaemenids, the Persians were 

86 Trans by A. Leo Oppenheim, with the Pyankov, "Borba Kira II s Astiagom," VDl, no. 3 

lacunae in the Akkadian text shown by brackets (1971), 16-37. Some of his conclusions, e.g., p. 23 

and additions of the translator in parentheses, in that Astyages came from Susa to Pasargadae to 

J. B. Pritchard, Ancient Near Eastern Texts fight Cyrus, are hardly acceptable. He also makes 

(Princeton, 1955), 305. little reference to the cuneiform sources. 

81 Ibid., 315. The archaic reference to Guti and 88 Herzfeld, op. cit. 1 [ch. 3, n. 31], 48; 

Manda-hordes (Ummanmanda) is characteristic of Dyakonov, Istoriya Midii [n. 4], 415-16. 

Babylonian sources. On the Harran text see S. 89 The fact that the Median rebel against Darius, 

Smith, Babylonian Historical Texts (London, 1924), in his Behistun inscription (II, 13, or par. 24) 

44. The best survey of Classical sources on the claimed to be of the family of Cyaxares and not of 

battle between Cyrus and Astyages is by I. V. Astyages is interesting in this regard. 



Medes, Scythians and Eastern Rulers 85 

exempt from taxes while the Medes paid as other subjects did (Herodotus III, 89, 97). 
Medes raised several revolts under Achaemenid rule, as we shall see, but the revolts 
may be attributed to the general oppression of the people by their rulers rather than to 
an ethnic antagonism. Gradually the Medes and Persians, as well as other Iranians, 
became mixed, which was in part a result of the world empire of the Achaemenids. 



CHAPTER V 
ACHAEMENIDS 



Literature: Our sources for the early Achaemenids are more abundant than for any period of the early 
history of Iran until the Sasanian period. In the last four decades the results of the excavations at Persepolis 
have enormously enriched our knowledge of the Achaemenid court. The monumental three volumes by 
E. Schmidt, Persepolis (Chicago, 1 953, 1 957, 1970) are the basic handbook for the site, but the epigraphical 
finds of Elamite tablets and Aramaic inscriptions have produced three books, G. Cameron, Persepolis 
Treasury Tablets (Chicago, 1948), R. T. Hallock, Persepolis Fortification Tablets (Chicago, 1969), and R. A. 
Bowman, Aramaic Ritual Texts from Persepolis (Chicago, 1970). The articles which these works have 
engendered, plus new Elamite tablets, would require an enormous bibliography. To select those especially 
noteworthy we may mention the articles, "Persepolis Treasury Tablets Old and New," and "New Tablets 
from the Persepolis Treasury," by Cameron in JNES, 17 (1958), 161-76 and JNES, 24 (1965), 167-92, 
and especially Hallock's "The Evidence of the Persepolis Tablets," The Cambridge History of Iran, 2 
(Cambridge University, Middle East Centre separate publication, 1971), 31 pp. Many articles have been 
written elucidating various details of the tablets. For a bibliography of articles up to 1970 see W. Hinz, 
'Die Quellen,' in G. Walser, ed., "Beitra'ge zur Acha'menidengeschichte," in Historia, 1 8 (1 972), 5-1 4. Since 
this publication, the following can be added for the Elamite tablets: I. Gershevitch, "Iranian Nouns and 
Names in Elamite Garb," TPS (1969-70), 165-99,, his "Amber at Persepolis," in Studia Classica et 
Orientalia Oblata, 2 (Rome, 1969), 167-251, "Island-Bay and the Lion," BSOAS, 33 (1970), 82-91, and 
"The Alloglottography of Old Persian," TPS (1979), 114-90. The many publications of M. Mayrhofer, 
especially in the realm of onomastica, fortunately are listed in his Ausgewihlte Kleine Schriften (Vienna, 
1978), and all Iranian names will be treated in the various fascicules of his forthcoming Iranisches 
Personennamenbuch, but until then see his Onomastica Persepolitana (Vienna, 1973), 358 pp. For a 
bibliography of Old Persian see Mayrhofer's, "Das Altpersische seit 1964," in W. B. Henning Memorial 
Volume, ed. by M. Boyce and I. Gershevitch (London, 1970), 276-98, and succeeding notices in the list of 
his own publications above. Further Hinz's Wege [ch. 4, n. 66], 174 pp. deals with the Elamite tablets as 
does his article, "Achamenidische Hofverwaltung," ZA, 61 (1971), 260-311, and "Die elamischen 
Buchungstafelchen der Darius-Zeit," Orientalia, 39 (1970), 421-40, and finally his Altiranisches Sprachgut 
der NebenOberlieferungen (Wiesbaden, 1975), 303 pp. 

A series of articles by R. Schmitt mostly regarding Iranian terms and names in the Elamite tablets have 
appeared; for a bibliography of his articles and others relating to philological questions, for the most part, 
see the bibliography in M. A. Dandamaev, Persien unter den ersten AchUmeniden (Wiesbaden, 1976), 257- 
62. 

For the Aramaic inscriptions on the green chert mortars, pestles and plates from Persepolis a large 
literature exists, among which we may mention P. Bernard, "Les mortiers et pilons inscrits de Persepolis," 
SI, 1 (1972), 165-76; B. A. Levine, "Aramaic Texts from Persepolis," JAOS, 92 (1972), 70-79; J. Naveh 
and S. Shaked, "Ritual Texts or Treasury Documents," Orientalia, 42 (1 973), 445-57; J. A. Delaunay, "A 
propos des Aramaic Ritual Texts," Al, 2 (1974), 193-21 7, and K. Kamioka, "Philological Observations on 
the Aramaic texts from Persepolis," Orient, 1 1 (Tokyo, 1975), 45-66. Most scholars now agree that the 
texts on the objects are not ritual texts but are records of the manufacture of the objects and their 
presentation to Achaemenid officials; further they were made in Arachosia (modern Qandahar area). The 
purpose of the mortars and pestles was hardly just to press grapes or grind spices, but was probably for the 
haoma ritual, part of the ancient Indo-Iranian religion, and apparently practiced by some Iranians in the 
Persepolis area. 

One of the most important new inscriptional sources for early Achaemenid history was the discovery 
of the quadri-lingual (OP, Akkadian, Elamite and hieroglyphic Egyptian) inscription on the statue of 
Darius I in Susa in 1973, and an entire issue of the Cahiers de la delegation archeologique francaise en Iran, 4 
(Paris, 1974) is devoted to the statue, as well as articles in J A (1 972). Other discoveries of new inscriptions 
were merely copies of inscriptions already known; cf. B. Gharib, "A Newly Found Old Persion 
Inscription," IA, 8 (1968), 54-69, and W. Hinz, "Eine neue Xerxes-Inschrift aus Persepolis," in his book 
Altiranische Funde und Forschungen (Berlin, 1969), 45-51. 



88 Chapter V 

Perhaps the most important epigraphical material for Achaemenid history, however, comes from 
inscriptions in Aramaic and in Greek found outside of Iran, but giving valuable information about the 
administration of the empire. It is, of course, impossible to list the numerous publications on Aramaic, but 
a bibliography by J. Teixidor appears regularly in the journal Syria called "Bulletin depigraphie 
Semitique." The ongoing Neue Ephemerisfilr Semitische Epigraphik by R. Degen (Wiesbaden, 1972-) is a 
handbook of Aramaic inscriptions to be consulted for all Aramaic inscriptions. Of special importance are 
the Aramaic papyri from Egypt of the satrap Arsames under Artaxerxes I and Darius II published by 
G. R. Driver, Aramaic Documents of the Fifth Century B.C. (Oxford, 1957), and E. G. Kraeling, The 
Brooklyn Museum Aramaic Papyri (New Haven, 1953), which supplement the old collection by A. 
Cowley, Aramaic Papyri of the Fifth Century B.C. (Oxford, 1923) ; cf. B. Porten, Archivesfrom Elephantine 
(University of California Press, 1968) and J. Naveh and S. Shaked, "A Recently Published Aramaic 
Papyrus (from Berlin)," JAOS, 91 (1971), 379-82, as well as P. Grelot, Documents aramtens d'Egypte 
(Paris, 1972). The Aramaic sources give us Achaemenid titles and names as well as other information. 

Excavations in Anatolia, notably at Xanthos in Lycia and at Sardis, have given us new Greek 
epigraphical remains from the Achaemenid period. Especially interesting is the trilingual inscription of 
Pixodaros, satrap of Lycia, in Greek, Lycian and Aramaic; see the CRAI (1974), 82-93, "Le stele trilingue 
recemment Ai couverte au Letoon de Xanthos," by H. Metzger, to be published in full in vol. 6 of Fouilles 
de Xanthos. For Sardis see L. Robert, Nouvelles inscriptions de Sardis (Paris, 1964). The fragmentary Greek 
inscriptions from Persepolis unfortunately add little to Achaemenid history; cf. G. P. Carratelli, "Greek 
Inscriptions of the Middle East, Persepolis," EW, 16 (1966), 31-34. They are probably marks by masons. 

Numismatics and inscribed seals enter the repertoire of sources on the Achaemenids. A survey of the 
coinage may be found in D. Schlumberger, V argent grec dans {'empire achimenide, in MDAFA, 14 (1953), 
55-57. Since that time we have S. P. Noe, Two Hoards of Persian Sigloi, NNM, 1 36 (1956). Further articles 
on satrapal issues may be found in the excellent yearly bibliography Numismatic Literature published by 
the ANS in New York. The far-flung circulation of coinage with Greek legends in the Achaemenid 
Empire is now assured. Achaemenid seals and sealings are also a source for history, and the continuity of 
earlier commercial practices into the Achaemenid period is revealed by the clay sealings. On them see my 
"The Use of Clay Sealings in Sasanian Iran," AI, Varia 1976, 5 (Leiden, 1977), 117-24, for general 
remarks on earlier sealings. The article by C.J. Gadd, "Achaemenid Seals," SPA, 1 , 383-88, is still valuable 
as are the articles by G. M. A. Richter and H. Seyrig in Archaeologica Orientalia in Memoriam Ernst 
Herzfeld, ed. by G. Miles (New York, 1952). The seals and sealings, especially those from Persepolis, 
provide material for reconstructing the material culture of the Achaemenids as well as their artistic 
achievements. 

Archaeology, of course, provides the basis for the dating and attribution of most aspects of material 
culture, and in addition to the volumes of Schmidt on the excavations at Persepolis mentioned above, the 
two volumes of Ann Britt Tilia, Studies and Restorations at Persepolis and Other Sites of Fars, ISMEO 
(Roma, 1 972 and 1 978), and the book by D. Stronach, Pasargadae (Oxford, 1 978), 312 pp., are significant. 
Work at Susa, the winter capital of the Achaemenid kings, continues, and results are published in the 
Cahiers of the French delegation in Iran and elsewhere. Fortunately, a classified and detailed yearly 
bibliography of the archaeology of Iran appears in AMI. 

Journals which contain articles on the Achaemenids or related Iranian subjects are lran,JNES, IA, SI, 
VDI, Persica, EW and Orientalia, while the various general Oriental journals published in different 
countries, all must be surveyed for articles relating to the Achaemenids. In addition, Archaeology, 
Archeologia Viva, Iraq, Syria, Sumer, Afghanistan, Anatolia, Mesopotamia, and Sovetskaya Arkheologiya 
from time to time have articles relating to the Achaemenids. Especially noteworthy was the series Acta 
Iranica (AI), published in Leiden, and the yearly reports on excavations in Iran, published in Tehran, plus 
the proceedings of the International Congress of Iranian Art and Archaeology. 

The Classical sources for the Achaemenids are well known: Herodotus, Xenophon, especially his 
Anabasis, Ctesias, the Alexander historians, especially Arrian, Curtius, Diodorus Siculus, together with 
Plutarch's life of Artaxerxes, Aeschylus, The Persians, Ptolemy for historical geography, and, of course, 
later authors such as Strabo, Pliny, as well as fragments of minor Greek writers. The Old Testament (esp. 
Esther, Ezra and Nehemiah) also provides some information about the Achaemenids, as do late fourth 
century Aramaic papyri and coins from Palestine; see P. W. Lapp, Discoveries in the Wadt ed-D&liyeh, 
Annual of the American Schools of Oriental Research (Cambridge, Mass., 1974), 17-30. At any time the 
spades of archaeologists may uncover material which will greatly change our views of Achaemenid 
history, although on the whole we may expect confirmation of what the literary sources we now have tell 
us. (Additional bibliography will be found in the footnotes.) 



Achaemenids 89 

CYRUS AND THE FOUNDATION OF THE EMPIRE 

We have already discussed briefly the Cyrus legend and the Parsua, but a few 
additional remarks about Fars before Cyrus may help elucidate the circumstances of 
the founding of his empire. The question of whether after the Landaufnahme by the 
Persian tribes in Fars, rule was divided into two collateral families of children of 
Achaemenes has been proposed by many scholars on the basis of the split genealogy of 
the Achaemenids after Teispes. It is more likely, however, given the nature of tribal 
societies and the broken mountains and valleys of Fars, that even after Teispes not just 
two but many tribal chiefs or 'kings' existed in Fars, and it was probably only under 
Cyrus the Great that they were all united under him as head of the Achaemenid clan 
of the Pasargadae tribe. Whether the process of unification followed much the same 
pattern as later under Ardashir, founder of the Sasanian dynasty, is, of course, 
unknown but not improbable. 

Although Xenophon (Cyropaedia, I, 2, 5) says the Persians were divided into 
twelve tribes, which number appears for other purposes in Xenophon, one should 
rather follow Herodotus (1, 1 25), who says there were many tribes, but he only names 
ten, including the Pasargadae, Maraphi, and Maspi. The first presumably lived 
around the capital of Cyrus and we may presume that the site took its name from the 
tribe. 1 The second had several prominent representatives mentioned by Herodotus 
(IV, 167; see also Aeschylus, The Persians, 778) attesting the reality of the tribe, while 
the third, according to Herzfeld, is preserved in a tribal name of the Lurs at the present 
time. 2 The name is also conceivably to be found in the province of Massabatike or 
Messabate between Khuzistan and Fars. 3 The other tribes mentioned by Herodotus 
are more problematic, for of those he calls settled and agriculturists, the Panthialai, 
Derousiai and Germani, only the last can be located in present Kerman. Herzfeld 
suggests that the first have their name preserved in the modern town of Fahliyan, but 
his etymology is unconvincing, even though the area must have been settled by one of 
the Persian tribes. 4 The Derousiai are not mentioned elsewhere, and it is not possible 
to locate them. For want of anyone in the present Fasa-Darab-Jahrum area of today, 
this tribe might be consigned to somewhere in that expanse, but this is mere surmise. 

The rest of the tribes Herodotus calls nomadic Dahi, Mardi, Dropiki and Sagarti, 
all of whom are found elsewhere on the plateau, or rather they are primarily 
elsewhere rather than solely in Fars. This argues for the correctness of the historian's 
assertion that they were nomadic, and also for the possible Persian-Central Asian 
(Sogdian) linguistic connections. The first are mentioned by many Classical authors 
(see PW) as living primarily east of the Caspian Sea, and the Dropiki or Derbikes are 

1 Many etymologies of the name, both ancient 2 E. Herzfeld, Zoroaster, 2 [ch. 3, n. 31], 729. 

and modern, have been proposed, most of them 3 For references see Mannert, op. cit. 2 [supra, ch. 

reviewed by Stronach, Pasargadae (Oxford, 1978), 1 , p. 1 ], 356. 

280-82. Perhaps the most plausible is by Bailey: 4 Herzfeld, Zoroaster, 2 [ch. 3, n. 31], 731. The 

'the Persian settlement.' (in D. Stronach, op. cit. 281, etymologies proposed by him in The Persian 

footnote 17) W. Hinz, Darius und die Perser Empire, ed. by G. Walser (Wiesbaden, 1968), 298, 

(Baden-Baden, 1976), 58, says the word was are interesting but unconvincing. 
Median in origin *Patragada, meaning "robbers' 
nest," not a likely name for a tribe. 



90 Chapter V 

located there as well. The Sagartians, who probably lived on the edges of the central 
desert near Yazd, were said by Herodotus (VII, 85) to be Persian in speech. The name 
is found in the Behistun inscription (II, 79), and a revolting Sagartian claimed to be 
descended from the Median kings. The Mardi present problems since they are found 
in so many places (see PW) in Armenia, Hyrcania and Media as well as in Fars. There 
is no compelling reason to call them a clan instead of a tribe and equate the name with 
the Maraphi as Herzfeld does. 5 Another presumably tribal name rather than a 
geographical designation, which Strabo (XV, 727) mentions, is Patishuvari(i) on the 
tomb of Darius, showing a spear bearer called Gobryas, as being a Pateisxoreis. 
Nothing else is known about them except the speculation among modern scholars 
that they were not a tribe but a clan, or that the name is the same as Pasargadae. 6 One 
may count the various tribes as part of the Persian folk or people, a branch of the 
Iranians, who settled in the south and who gradually assimilated with the pre-Iranian 
settled population. As usual, the tribes lost their importance in the course of settlement 
and the extended family, or clan, and the large unit of the 'people,' now identified 
with the land they occupied, became important. This is not to deny the existence of 
tribal areas unsuited for other than a pastoral life, but history was made primarily by 
the settled folk, without whom the pastoralists hardly could have existed. It was in 
such a situation, the settling of the Persian tribes and their mixing with the original 
population, that Cyrus was able to unite his people and lead them to victory over the 
Medes. 7 

We have already mentioned the Nabonidus chronicle which tells of the defeat and 
capture of Astyages by Cyrus. Whatever the facts about the desertion of the Median 
army to Cyrus, the latter must have been in a strong position to challenge his lord. 
One question which arises is the extent and power of vassal rulers in the Median 
Empire and in Fars before Cyrus. Cyrus in an Akkadian inscription says, "I am Cyrus, 
king of the world, great king, legitimate king, king of Babylon, king of Sumer and 
Akkad, king of the four rims (of the earth), son of Cambyses, great king, king of 
Anshan, grandson of Cyrus, great king, king of Anshan, descendant of Teispes, great 
king, king of Anshan, of a family (which) always (exercised) kingship." 8 If we 
compare this with the statement of Darius (Behistun I, 6-11) that from long ago his 
family had been kings; eight had been kings before him, then obviously we can accept 
two lines of kings in Fars. Probably the Achaemenids ruled the most productive parts 
of the province — the two plains where Persepolis and Pasargadae were located. Did 
other 'kings' rule in the smaller oases of present Darab, Fasa, Shiraz and others? 
Presumably the chiefs of the various Persian tribes in their expansion over the 

5 Herzfeld, op. cit. 1 [ch. 3, n. 31], 70. Hinz, op. Arlaxerxes I, 2), an Elamite name, or Kuru, an 
cit. [n. 1], 104, claims that the Kurds are ancient Indian name meaning 'blind.' Cf. in detail 
descendants of the Sagartians, which is purely W. Eilers, "Kyros, eine namenkundliche Studie," 
guesswork. BZN, 1 5 (1 964), 1 80-236. J. Harmatta, "The Rise 

6 Cf. R. H. von Gall, "Persische und Medische of the Old Persian Empire," AA, 19 (1971), 7-8, 
Sta'mme," Proceedings of the 1st Annual Symposium criticized the various theories and convincingly 
of Archaeological Research in Iran (Tehran, 1972), 2. claimed all early Achaemenid royal names could 

7 The name of Cyrus has received many be explained as Iranian. His further assertion that 
etymologies from 'son' as Luri kur proposed by they are eastern (not western) Iranian names, 
V. I. Abaev, "K etymologii drevnepersidskikh however, makes little sense for that early period, 
imen," Etimologiya, 1965 (Moscow, 1967), 289- 8 L. Oppenheim's translation on p. 316 of 
91, to 'fire' or 'the sun' (Ctesias in Plutarch's Pritchard, op. cit. [ch. 4, n. 86]. 



Achaemenids 91 

province did maintain separate domains until Teispes divided the age old center of 
Pars, Anshan, between his sons Cyrus and Ariaramnes. The Achaemenid kings 
apparently were long-lived as well as capable since they brought all Persians under 
their banner, just how or when is impossible to say. 9 Possibly the Median system of 
rule favored a pyramidical structure of the allegiance of small rulers through a 
provincial king to the central court in Ecbatana, a nascent satrapal and feudal system. 
Harmatta, on the basis of presumed Median titles and bureaucratic offices in the Old 
Persian inscriptions, plus the passage in Behistun (1, 64—5) where Darius restored what 
Gaumata supposedly destroyed, concluded that the Median state was a more 
centralized state than the Achaemenid empire. 10 He further asserts that the 
Achaemenid kings established military fiefs in royal domains and gave them as 
payment for military services to the generals of their armies, whereas the Median 
kings had direct bureaucratic control over their royal lands. This is a plausible theory 
if by no means stated in any source, but in any case one might expect the Medes to 
have developed a more imperial government structure than the Persians at this time. 
We will examine briefly the question of military fiefs below. Suffice it to say that the 
Persian tribes rallied around Cyrus, whereas some of the Medes deserted their king. 
Whether those who deserted to Cyrus were disgruntled aristocrats who opposed the 
centralizing of power in the hands of Astyages is pure speculation. 

The length of time of hostilities between Medes and Persians is difficult to 
determine since Herodotus (I, 127-28) implies two battles in the last of which 
Astyages was captured. The sequence of events may be constructed as follows: about 
559-558 B.C. Cyrus became king. Dandamaev claims he at once rebelled against the 
Medes, but this is unlikely. 11 Rather his early years probably were spent in 
consolidating his power in Fars. Two Akkadian sources contradict each other on the 
chronology. One, the 'Sippar' cylinder, about the repair of the temple to Sin the moon 
god in Harran by Nabonidus, said, "When the third year came to pass, he (Marduk, 
god of Babylon) made rise against them (Umman-manda or Medes), Cyrus king of 
Anshan, his young servant, and he (Cyrus) scattered the numerous Ummdn-manda 
with his small army and captured Astyages, king of the Umman-manda and brought 
him in fetters into his (Cyrus') land." 12 The 'Nabonidus' chronicle, however, claimed 
this took place in the sixth year of Nabonidus. Tadmor has convincingly 
demonstrated that one should accept the date given by the Nabonidus chronicle and 
not the 'dream' text of Sippar, since such 'dream' texts, as well as poetry, had different 
objectives than a chronicle. 13 So we may date the overthrow and capture of Astyages 

9 If Cyrus I is the ruler who sent his son as a ' 2 Translation in S. Smith, Babylonian Historical 
hostage to the Assyrians c. 639 B.C. then Cambyses Texts, 44, rev. by L. Oppenheim in H. Tadmor, 
I must have had a long life, given the dates of Cyrus "The Inscriptions of Nabunaid: Historical Ar- 
il. Furthermore, Xerxes (XPf, 15-27) says that rangement," Assyriological Studies, 16 (Chicago, 
both his grandfather Hystaspes and his greatgrand- 1965), 351. 

father Arsames were alive when his father Darius l3 Tadmor, op. cit. [n. 12], 352-54. The verse 

became king! It is by no means impossible for account of the same time period, with many 

several generations to span a large extent of time, lacunae, trans, by S. Smith, Babylonian Historical 

but his statement is nonetheless remarkable. Texts, 86-91 , and in Pritchard, op. cit. [ch. 4, n. 86], 

10 J. Harmatta, "The Rise of the Old Persian 312-15, cannot be used for chronology. 
Empire", AA, 19 (1971), 14-15. 

M. Dandamaev, "Politische und wirtschaft- 
liche Geschichte," Historia, 18 (1972), 16. 



92 Chapter V 

at 550/549 B.C., but we still do not know when hostilities began or how long they 
lasted. Although frequently asserted, there is no evidence that Cyrus made an alliance 
with Nabonidus, ruler of Babylonia, against Astyages. If an alliance were made by 
Cyrus with anyone against Astyages, it would have been more likely with other 
Iranian peoples or tribes than with the Babylonians, although the accounts of Classical 
authors are not reliable in this matter. 14 What is certain is that Cyrus plundered the 
treasury of Ecbatana and took the spoils to Fars, perhaps to his new capital at 
Pasargadae. The fate of Astyages according to Herodotus (I, 130) was to live with. 
Cyrus until he died, or he was banished and accidentally killed later (Ctesias, para 5). 
Neither the Cyrus saga, Xenophon's Cyropaedia, nor any source, says that Astyages 
was executed by Cyrus, so we may suppose he was treated well. After this conquest 
Cyrus had to insure fealty of his new Median subjects to himself and the Persians 
(Justin 1, 7, 2). Whether at this time Cyrus undertook expeditions to the east beyond 
the borders of former Median control is unknown but not improbable. What we do 
know, however, is the outcome of the war with Lydia in the west. 

According to the 'Nabonidus Chronicle' in the ninth year of his reign (547 B.C.), "in 
Nisan (April-May) Cyrus, king of Persis, called up his army and crossed the Tigris 
below the town Arbela. In the month Ayar (May-June) he marched against the 
country Lu . . . killed its king, took his possessions, put (there) a garrison of his own. 
Afterwards, his garrison as well as the king remained there." Much controversy has 
raged about the identity of the country name beginning with Lu-, many scholars 
suggesting Lydia, in spite of the testimony of Classical authors that the king of Lydia 
was not killed by Cyrus. 15 Hinz proposed to read Su- and reconstruct Suhi, a district 
on the middle Euphrates, but this emendation is not generally accepted. 16 Another 
suggestion, that the word for 'kill' (col. II, 17-iduk) can also mean 'beat' or 'crush' 
militarily, would bring the Akkadian source in line with the other sources. Again, we 
have only Herodotus (I, 75-91), among the Classical writers, who gives a detailed 
account of the fall of Lydia to Cyrus. According to the historian, hostilities were 
opened by Croesus, the Lydian king, by crossing the Halys River which had been the 
boundary with the Median state to the east. Cyrus marched to meet the Lydians, but 
in a hard fought battle neither side was victorious. Croesus went into winter quarters 
at his capital Sardis with the intention of renewing the war in the spring with the aid 
of the Egyptians and Ionians. Cyrus, however, did not wait but came to Sardis and 
after a skirmish besieged Croesus in his capital and after fourteen days captured it. At 
first Cyrus prepared a pyre for the defeated monarch but after lighting it he changed 
his mind; the fire was extinguished and Croesus was taken as a prisoner into Iran. 



14 See the discussion in I. V. Pyankov, "Borba Chronicle and the Fall of Lydia," AJAH, 2 
Kira s Astiagom," VDI (1971), part 3, 16-36 and (1978), 97-116, with no definite conclusion. 
83-93. ' 6 Hinz, op. cit. [n. 1 ], 97. The exact location of 

15 According to Dandamaev, supra, Persien, 95, Suhi is uncertain, but presumably it was in some 
"Laut babylonischen Quellen sei Kroisos mit dem way either tributary to or under the rule of 
Tode bestraft worden." His plural is unwarranted, Babylonia. Cyrus would have crossed the Tigris to 
for the only Babylonian text which is relevant is the go west in any case, whether against Suhi or Lydia, 
one under discussion. The Greek texts are Hero- so neither that sentence nor the date (usually 
dotus, Ctesias and their successors. On the reading interpreted as 547 or 546 B.C.) helps us to decide 
of the chronicle see J. Cargill, "The Nabonidus exactly where Cyrus went. 



Achaemenids 93 

Greek vases showing Croesus on the pyre do not contradict the account of Herodotus, 
as implied by some scholars. 17 

Much has been written about the first contacts of the Persians with the Ionian 
Greeks, but Cyrus clearly did not put much weight on their importance after the fall 
of Sardis, for there is no reason to doubt Herodotus (1, 153) who says that Cyrus left 
the conquest of the coastal Ionian cities to a general, while he himself, with Croesus, 
went to Ecbatana, to prepare for the conquest of Babylonia, the Bactrians, Sakas, and 
Egypt, which problems were more on his mind. The Ionian cities, as well as Caria, 
Lycia and Phrygia, were conquered by two Median generals, first Mazares and, after 
his death, Harpagos, according to Herodotus. The Greco-Persian wars, and relations 
between the two, were long regarded as a conflict between democracy and autocracy, 
but this simplistic view has long been abandoned. It is not the purpose of this work to 
repeat what H. Bengtson has already described, but some of the events seen from the 
Persian rather than the Greek side may change our traditional views of history based 
on both sources and prejudices from one side alone. 18 First, changes in the 
administration of the far west of the Achaemenid Empire were not made overnight, 
but were developed over much time; most of them, as we shall see, were initiated by 
Darius and his successors. Second, Anatolia was not Greek territory, and the Ionians 
themselves were invaders and colonizers, not always a boon and blessing to the 
natives as frequently has been proclaimed. Third, the vacillations and changes, 
alliances and policies of the Ionian coastal cities, frequently the result of extensive 
bribery, do not present an edifying picture of the Greeks. Just as all Greeks did not 
follow Athens, so not all Achaemenid governors or satraps did what the court at 
Persepolis ordered. This does not mean that Greeks were satisfied under Persian rule, 
any more than were the Egyptians or others, but the relative loyalty of the Ionian 
Greeks compared, for example, to the Egyptians, to Achaemenid rule bears inves- 
tigation, which we will attempt to explain after a survey of the later fortunes of the 
empire. 

Cyrus probably turned his attention to the east after the conquest of Lydia, in the 
years 546-539 B.C. Herodotus (I, 177-78) says that while Harpagos ravaged the 
'lower' parts of Asia, Cyrus conquered 'upper' Asia, all of its peoples and not except- 
ing any, after which Cyrus turned his attention to the Assyrians (meaning the 
Babylonians). This usually has been interpreted to mean that Cyrus conquered the 
eastern regions before his advance on Babylon, but Khlopin has argued that the words 
of Herodotus only refer to Asia Minor and not to the east. 19 On the basis of the 



17 Xenophon, Cyropaedi'a, VII, 2, 3, has nothing 18 H. Bengtson, Griechische Geschichtes (Mu- 

about the pyre nor does Ctesias, 4. All agree that nich, 1977), 130-31. The interpretation of Greek 

Croesus was well treated by Cyrus. The scenes on actions as part of a great power struggle between 

the amphorae, however, show Croesus immolat- Carthaginians in the west and Persians in the east 

ing himself, probably merely following a canon of with the Greeks defending liberty between the 

art. Cf. J. D. Beazley, Attic Red-Figure Vase-Painters two is an example of imposing modern concepts 

(Oxford, 1963), 238; on a poem regarding this on the past thereby distorting our understanding, 

scene see S. Smith, "Illustrations to Bacchylides," ,9 I. N. Khlopin, "Baktriiskii pokhod Kira II," 

Joi/rna/o/"He;/e«iV:SfM</iei,18(1898),267,andJ.G. AF, 1 (1974), 208. 
Pedley, Ancient Literary Sources on Sardis (Cam- 
bridge, Mass., 1972), 41. 



y* Chapter V 

revolts against Darius mentioned in the Behistun inscription, he further claims that 
only Parthia and Hyrcania had been under Median rule, since both areas supported 
the Median rebel against Darius while Bactria, Arachosia, Sogdia, and other areas in 
the east remained true to Darius, evidence of their prior submission to Cyrus as part of 
a Bactrian kingdom conquered by Cyrus. Again, the truth cannot be determined and 
other sources, such as Xenophon, Berossus, or Justin do not help in this regard. 20 No 
one disputes the conquests of Cyrus in the east and in Central Asia, because of the 
town of Kureskhata (Latin: Cyreschata), medieval Kurkath, and the story of the 
Arimaspi in Central Asia as supporters of Cyrus, but the question is when did he 
conquer them? 21 Since the conquest of Babylon took place in October 539 B.C. it is 
not easy to believe that Cyrus spent several years without any activity simply resting 
in palaces on the Iranian plateau. Just how much territory he conquered in the east 
before Babylon and how much just before his death unfortunately cannot be 
determined. 

The fall of Babylon was an important event in world history. Many inscriptions of 
Nabonidus, last king of Babylonia, are known and a fairly detailed account of his stay 
in the north Arabian oasis of Taima, until about 540 B.C., his building activities, and 
reconstruction of the temple to the deity Sin in Harran, can now be reconstructed. 
The Babylonians were unhappy with their ruler and in the famous cuneiform 
'Nabonidus' chronicle we hear the story: "In the month Tashritu when Cyrus 
attacked the army of Akad in Opis on the Tigris, the inhabitants of Akkad revolted, 
but he (Nabonidus) massacred the confused inhabitants. The fourteenth day, Sippar 
was seized, without battle. Nabonidus fled. The sixteenth day, Ugbaru governor of 
Gutium (hodie west Kurdistan) and the army of Cyrus entered Babylon without 

battle. Afterwards Nabonidus was arrested in Babylon when he returned (there) 

In the month of Arahsammu, the third day (29 October 539 B.C.) Cyrus entered 
Babylon ; green twigs were spread in front of him — the state of 'peace' was imposed on 
the city." 22 Akkadian sources tell us how Nabonidus did not come to Babylon for the 
new year's festival in honor of Marduk until the year before the fall of the city. The 
great 'Median wall' between the two rivers at Sippar did not halt the invaders, and 
many captive peoples in Babylon looked upon Cyrus as a deliverer. Certainly Isaiah 
(ch. 35, 40-55) and Ezra (ch. 1) testify to the high regard held for the Persian 
conqueror by some of his new subjects. Cyrus regarded himself as a legitimate ruler of 
Babylonia and performed ritual acts in a temple of Marduk to conciliate the citizenry, 
and if we are to believe the cuneiform sources he did just that. These sources, however, 

20 Contrary to Dandamaev, supra, Persien, 96. (1947), 163-66 (also Strabo, 517) and for the 

The use of the word ko.tw 'lower' or 'seaward' Arimaspi see Diod, XVII, 81, also in Arnan VI, 24, 

Asia, is ambiguous; in Herodotus it would seem to Curtius (VII, 3), etc., although their location is 

refer to the coastal areas of Asia Minor, while the uncertain. Accounts of the meeting between Cyrus 

rest of Asia would be the east, and not just the and Zoroaster in eastern Iran, as several scholars 

mountainous areas of Anatolia as opposed to the have proposed, is only fascinating fantasy, 
coast. Earlier, the horizons of the Greeks may have 22 Pntchard, op. cit. [ch. 4, n. 86], 306. The 

been limited such that Asia meant only Anatolia, Cyrus cylinder (ibid., 315-16) has much the same, 

but by the time of Herodotus (e.g., I, 4) Asia with many praises for Cyrus and his close 

represented much more. relationship with Marduk the god of Babylon. 

On the city (today Ura-tyube) cf. references The new inscriptions of Nabonidus from Taima 

m E. Benveniste, "La ville de Cyreschata," JA relate to his stay there. 



Achaemenids 95 

may be propaganda documents for Cyrus; the account in Herodotus (1, 189-191) tells 
how Cyrus after a long siege changed the course of the river (Euphrates, sic) and 
entered the city by a trick, a good story but unacceptable as history. 

After the conquest of Babylon, "all the kings of the entire world from the Upper 
(Mediterranean) to the lower (Persian Gulf) Sea, those who are seated in throne 
rooms, (those who) live in [other types of building as well as] all the kings of the West 
land living in tents, brought their heavy tributes and kissed my feet in Babylon." 23 
Thus, Cyrus became the heir of the Babylonian kings in the allegiance of the peoples 
of the 'Fertile Crescent' to the borders of Egypt. From Akkadian business documents it 
is clear that Cyrus introduced no great changes in the economy or the practices of his 
new subject peoples, and no accounts of rebellions or resistance to Cyrus have been 
found, rather local officials, prices, etc. continued much as before Cyrus. 24 From 
Akkadian sources Dandamaev has concluded that Cambyses, son of Cyrus, was made 
king of Babylon shortly after the conquest, but then was removed and finally 
reinstated about 530 B.C. shortly before the last campaign and death of Cyrus in the 
east. 25 Undoubtedly in the years after the fall of Babylon, Cyrus was involved in the 
consolidation of the 'Fertile Crescent,' as well as other parts of the empire before an 
attack on Egypt. Because of the raids of nomads against the settlements in the 
northeast of his empire Cyrus had to undertake an expedition against the Massagetai, 
who roamed east of the Caspian Sea. In a battle against them he lost his life in the 
summer of 530 b.c. 26 The 'Cyrus Saga,' of course, relates the events leading to his 
death, but doubtless many versions of his end existed. According to Herodotus (I, 
205-15) Cyrus crossed the Araxes River to fight the queen of his opponents. Much 
has been written about this river, some claiming it to be the river of the same name in 
Transcaucasia, but Pyankov in a series of articles convincingly identified it as the 
lower Oxus by the Aral Sea, and he has elucidated the various versions of the war. 27 
First victorious, Cyrus was subsequently defeated and slain, but his body presumably 
was secured by the Persians and brought to Pasargadae for burial. If we accept 
Herodotus (1, 214) that Cyrus had reigned twenty-nine years, he was about sixty-nine 
when he died, and Cambyses succeeded him, with no opposition that has been 
recorded, at the end of August 530 B.C. according to Akkadian documents. 

Cyrus has an excellent reputation in history, apparently well deserved. 28 It is 
generally agreed that Pasargadae was his capital, for this is the only site which has 
inscriptions with his name. The inscriptions there have raised many problems about 

Ibid., 316. several scholars have proposed ; see W. B. Henning, 

24 Cf. Dandamaev, supra, Persien, 99. Zoroaster [ch. 3, n. 29], 23, n. 2. The name might be 

Ibid., 102, and note 406 with sources. It is interpreted as 'great Getae'. 

quite probable that Ugbaru, commander of the 27 I. V. Pyankov, "K Voprosu o marshrute 

army that entered Babylon, held the title 'king of pokhoda Kira II na Massagetov," VDI (1964), part 

Babylon' under Cyrus until his death a short time 3, 115-30; "Khorasmii Gekateya Miletskogo," 

after the conquest. Cf. W. H. Shea, "An unrecog- VDI (1972), part 2, 2-21, and "Massagety 

nized vassal king of Babylon in the early Gerodota," VDI (1975), part 2, 46-70. Less likely 

Achaemenid period," Andrews University Seminary is an identification of the river as the Jaxartes as 

Series, 9 (Berrien Sorings, Michigan, 1971), 113 does Hinz, op. cit. [n. 1], 118. 
foil. 



28 Herodotus (III, 89) writes the Persians say 
arius was a huckster, Can 
/rus a father," while Xenopl 
getai is not related to 'fish eaters,' or 'great Sakas,' as continuous eulogy of Cyrus. 



The date is from Akkadian documents; see "Darius was a huckster, Cambyses a despot but 
Dandamaev, supra, Persien, 103. The name Massa- Cyrus a father," while Xenophon's Cyropaedia is a 



96 Chapter V 

the origins of the Old Persian cuneiform writing system which has a large literature 
and many partisans of the view that the Old Persian system of writing existed long 
before Darius. 29 The linguistic, artistic, archaeological and historical arguments 
advanced by those scholars who support the antiquity of the system of writing would 
require volumes to elucidate, but the inscriptions of Cyrus at Pasargadae using the 
first person singular pronoun provide the primary textual basis for the position that 
the Old Persian cuneiform writing existed before Darius. Others are just as vehement 
in defending the thesis that Darius invented the system of writing, based primarily on 
a famous 'letter' of Themistocles and paragraph 70 of the Behistun inscription. 30 The 
'letter' mentions gold and silver vessels in Greece inscribed "with the old Assyrian 
letters, not those which Darius, father of Xerxes, recently wrote for the Persians." 31 
Without going into further discussion, a compromise between the two camps may be 
reached in proposing that Cyrus, or someone even earlier than he, may have ordered 
work begun on a system of writing for Iranian comparable to Akkadian and Elamite, 
but it came to fruition only under Darius who surely had more inscriptions written in 
it than anyone else, and he let this be known throughout the ancient world. 

To return to Pasargadae, archaeologists have determined that Cyrus did build an 
acropolis with massive stonework, and the basic ideas for the masonry probably came 
from Lydia and Ionia since masons' marks here are duplicates of those found at 
Sardis. 32 This would indicate a date for the building activity at Pasargadae after the 
conquest of Lydia. Cyrus' tomb and his palaces show similar influences from the west, 
although the end result is a unique blend, the beginning of Achaemenid art and 
architecture so often characterized as syncretic. While the remains at Pasargadae are at 
present not as impressive as those of Persepolis, with their original gardens and water 
courses Pasargadae bespeaks the character of Cyrus as reported in Classical sources, 
whereas Persepolis is ostentatious and more in character with Darius and his 
successors. 

CAMBYSES AND THE 'USURPATION' OF GAUMATA 

Cambyses is known in the sources for his conquest of Egypt, already planned during 
the lifetime of his father. 33 It already has been mentioned that, according to Akkadian 

29 More recent discussion on the pre-Darius writing system was introduced by Darius is W. 
origin of the system of writing are: I. M. Hmz; cf. especially his "Die Entstehung der 
Dyakonov, "The Origin of the 'Old Persian' altpersischen Schrift," AMI (1968), 95-98; also J. 
Writing System," in Boyce and Gershevitch, eds., Harmatta, The Bisutun Inscription and the 
supra, Henning Volume, 98-124, who argues for a Introduction of the Old Persian Cuneiform 
Median origin of the writing system. In this he is Script," AAH, 14 (1966), 255-83; C. Nylander, 
followed by R. Ghirshman, "A propos de lecriture "Who Wrote the Inscriptions at Pasargadae," 
cuneiforme vieux-perse," JNES, 24 (1965) 244- Orieulalia Seucana, 16 (1967), 135-80, and others. 
50,J. Kurylowicz,"Zur altpersischen Keilschrift," 3I C. Nylander, "AZEYPIA FPAMMATA, 
in his Esquisses Linguistiques (Cracow, 1960), 274- Opuscula Atheniensia VIII," Ada Insliluti Alhen- 
79 and with a detailed summary and bibliography iensis Regni Sueciae, 14 (Lund, 1968), 1 19-36. The 
by P. Le Coq, "Le probleme de lecriture Greek text is ((p'ols Ziriycy pa-mat to. 'Aoovpta 
cuneiforme vieux-perse," AI, 3 (1974), 25-107, to iraAaia ypap.pM.ra, oi>x a AapeTos 6 Trar-qp 
and esp. R. Schmitt, "I. Forschungsbericht, Altper- Eip£ov flepoais evayxos «?ypcu/ie. 

sische Forschung in den Siebzigerjahren," Kralylos, 32 Stronach, Pasargadae [supra, n. 1], 20-21. 

25 (1981), 1-66. "The name in OP, Kambujiya, has been 

30 The main partisan of the position that the explained as borrowed from a people on the Indian 



Achaemenids 91 

sources, it seems that Cambyses was made king of Babylonia by his father then 
removed and finally reinstated. His mother Kassandane, according to Herodotus (II, 
1), was the daughter of Pharnaspes, an Achaemenid (III, 2), and he had a brother 
Bardiya, whom the Greeks called Smerdis, plus several sisters, Atossa and Roxane, who 
died in Egypt (according to Ctesias, 12). Cambyses is supposed to have married his 
sisters, as well as Phaidyme daughter of Otanes, son of Pharnaspes (Herodotus III, 68). 
Two remarks may be made, first that succession under the Achaemenids did not 
follow the law of primogeniture but the ruler designated whomsoever of his sons he 
wanted to succeed him, and second, this is the first account among the Persians of 
next-of-kin or brother-sister marriage, which has aroused much controversy in 
Zoroastrian circles. 34 Next-of-kin marriage was fostered by later Zoroastrian priests 
but, according to Herodotus (HI, 31) Cambyses' desire to marry his sister was against 
the law, but the royal judges told him another law permitted the king to do what he 
wished, so he married his sisters. Since the Greeks regarded next-of-kin marriage as 
disgraceful and against the law, and since Herodotus gives a most hostile Greco- 
Egyptian, and perhaps a special Persian account of Cambyses, it is not unexpected that 
this sin should be laid upon the Achaemenid ruler. 35 Incest did and does occur, but 
among the Persians it is asserted that the practice was recommended as part of their 
religion. We shall examine this later under the Sasanians, but must be careful in 
ascribing a later religious injunction to the Achaemenid period. In any case, at that 
time the practice is not attested elsewhere. 

Early in 525 B.C. Cambyses invaded Egypt, several years after mounting the throne 
and after establishing his authority over the empire. We have Egyptian sources which 
do not corroborate the long and critical account of Herodotus (III, 1—38), rather it 
seems clear that Cambyses was accepted as a new pharaoh of the twenty-seventh 
dynasty, and he respected local traditions and honored the Egyptian religion. Like his 
father in Babylon, so Cambyses in Memphis kept the previous administration intact, 
only bringing Achaemenid garrisons into the land. 36 Foreign mercenaries such as the 
Jews of Elephantine Island in the upper Nile continued to serve now new masters. 

frontier, the Kambojas, in the same way as later trans. A. S. Way, 2 (London, 1965), 428-29, in a 

Roman appellatives such as 'Germanicus,' by J. speech of the Spartan Hermione to Andromache, 

Charpentier, "Der Name Kambyses," ZII, 2 an Oriental slave. The special Persian account 

(1923), 140-52, and followed by V. Abaev in about the misdeeds of the king is traced to the 

Elimologiya 1965 (Moscow, 1967), 291-92. See tribal Persian aristocracy, inimical to the central- 

also K. Hoffmann, "Vedische Namen," Worter und ized despotism of Cambyses, ace. to Dandamaev, 

Sachen, 21 (1940), 146. This explanation, how- supra, Persien, 156. 

ever, ignores the fact that there were at least two 36 The best account of this period is by G. P. 

rulers called Cambyses, and both are hardly Posener, La premiire domination perse en Egypte 

derived from the same source, while the activities (Cairo, 1936), followed faithfully by F. K. Kienitz, 

of Cambyses I in the east are both unattested and Die polilische Geschichte Agyptens vom 7. bis zum 4. 

""likely. Jh. vor der Zeilwende (Berlin, 1953). On the 

34 The practice of appointing the first son born Elephantine garrison see B. Porten, Archives from 
after his father became ruler did occur, but not as Elephantine, 421 pp. with bibliography. That 
an iron-clad rule. For next-of-kin marriages Cambyses dated his regnal years by the Egyptian 
among the rulers, the Elamite matriarchate may method and from the beginning of his reign, not 
have provided a model. his invasion of Egypt, has been shown by K. M. T. 

35 See H. Diels/W. Kranz, Fragmente der Vor- Atkinson, The Legitimacy of Cambyses and 
sokratiker, 2 (Berlin, 1951-52), 408, the Dissoi Darius as Kings of Egypt," JAOS, 76 (1956), 167- 
Logoi or Dialexis. See also Euripides, text and 70. 



98 Chapter V 

Expeditions up the Nile into Ethiopia and westward to the oasis of Amon were not 
successful according to Classical authors. The reasons for the later Egyptian 
denigration of Cambyses probably can best be explained by the reduction in revenues 
and land imposed on the temples of Egypt by Cambyses, and this action may have 
been provoked by an uprising in Egypt following his return from Ethiopia. 37 
Cambyses was also disparaged by the Greeks for his fratricide, and even in Plato's 
Laws (III, 694) the decline of the empire under Cambyses was attributed to the lack of 
discipline and pampering of princes when young. Further, according to Plato, 
Cambyses killed his brother through envy that his brother was equal to himself, and 
then through drunkenness and debauchery he lost his throne to the Medes, led by a 
eunuch who despised the stupidity of Cambyses. Plato probably gives a version 
current in Greece at the time, but this does not conflict with the account of Darius in 
his inscription on Behistun (I, 26-27). The account of Herodotus also does not 
conflict with the story given by the Behistun inscription, and to paraphrase the two, 
this is the combined account. Cambyses had a brother called Bardiya (Smerdis in 
Greek), both of the same mother. After he became king Cambyses murdered his 
brother but kept it a secret. When Cambyses was in Egypt a Magian called Gaumata 
(Herodotus says he had the same name as Bardiya) revolted and proclaimed himself 
Bardiya the true brother of Cambyses. The heart of the empire, Persis and Media, as 
well as other lands, supported the rebellion against Cambyses, but the latter died from 
a wound on the way back to Persis from Egypt. 38 According to Herodotus the new 
king was very popular since he instituted reforms, remitted taxes for three years, and 
in general was liked. He continues that only a few noble Persians suspected that the 
new king was in truth an imposter and he tells several stories about this discovery and 
the plot to overthrow him. The Behistun inscription is more laconic and simply 
relates that Darius with a few men attacked Gaumata and killed him with his chief 
followers. He then says that he restored property which had been confiscated and 
reestablished the people in their positions and over their property as it had been before 
the usurpation. 39 Afterwards he goes into great detail in the Behistun inscription 
about the various uprisings all over the eastern part of the empire which he had to 
suppress. Curiously nothing is said about rebellions in Egypt, Palestine or Anatolia in 
the inscription although they can hardly have remained completely loyal to Darius 
from the outset. 

In any case, such is the famous story of Darius' rise to power and the overthrow of 
pseudo-Bardiya, for both Herodotus and the Behistun inscription proclaim the 
usurpation of the Magian. Just how far one can interpret the struggle which brought 
Darius to power as a conflict of different interests or classes, or even ethnic differences 

37 Posener, op. cit. [n. 36], 170, n. 6, and Kienitz, 39 The translation in appendix 2 is mainly based 
op. cit. [n. 36], 59, with references. on R. Kent, Old Persian (New Haven, 1953), 119- 

38 Much has been written about this, but most 20. PaisiyJhuvJda has been identified as the present 
scholars agree it means he died a natural death district of Fasa in Fars by J. Hansman, "An 
(from an accident ace. to most Classical sources), Achaemenian Stronghold," Al, 6 (Monumentum 
and not suicide, or other. Dandamaev, supra, H. S. Nyberg 3) (1975), 304, while Herzfeld, 
Persien, 146-51, devotes a long, detailed discussion Zoroaster [ch. 3, n. 31], 1,205, identifies Sikayau- 
to this and concludes it meant Cambyses received vati with Sikawand near Kermanshah, which is 
his just dues' (accidental death) for his bad probably Sagvand in Luristan, an unlikely 
behavior. identification. 



Achaemenids 99 

rather than as a purely personal clash, is difficult to determine. Plato and others speak 
of the episode of the pseudo-Bardiya as an attempt to reassert Median preponderance 
over the Persians in the empire. It is curious in the history of Iran that when we have a 
primary source such as the Behistun inscription, many scholars do not believe the 
account of Darius but consider him a rebel against the true Bardiya. 40 The truth is 
that again we do not know and can prove nothing one way or the other. It would be 
impossible to even merely enumerate the pros and cons of both sides, but the partisans 
of Darius' mendacity cite the following arguments against him: (1) The accounts of 
Ctesias, Herodotus, Behistun and others have too many inconsistencies to be plausible 
and the story is too contrived to be believed; no one like the 'false' Bardiya could live 
for five or more years without being 'discovered.' The last argument is unconvincing, 
however, since no one claims that the Magian proclaimed himself to be Bardiya as 
soon as the latter was murdered. If he fooled people, it was only from the time of his 
revolt to his death, which as we know from Akkadian documents was from March to 
29 September 522 B.C. (2) Aeschylus (774-77) consideres Mardis/Mardos (Bardiya) 
the fifth legitimate king of the Achaemenids. (3) Herodotus (III, 72) claims Darius 
approved of lying, and in the Behistun inscription Darius speaks so much of 'the lie' 
that one is hard pressed to believe his integrity. Many more persuasive arguments 
have been advanced, none of which, however, 'proves' the position. The proponents 
of Darius' veracity assert that (1) The main sources should be believed unless proved 
to be false. (2) The second false-Bardiya in Persis after the death of Gaumata (Behistun 
III, 21-52), a certain Vahyazdata, caused much trouble, and he hardly would have also 
proclaimed himself Bardiya if something had not been wrong about the fate of the 
true Bardiya. (3) Comparisons with other false claimants in Iran's history such as the 
Safavid Isma'li II, are made by proponents of Darius' veracity, and there are also other 
arguments. 41 As remarked, not one of the arguments clinches the case one way or 
the other, but some details command more agreement than others. For example, the 
role of the Median Magi as opposed to Persians, or the slaughter of the Magi 
commemorated in a festival after the death of Gaumata, cannot be taken seriously or 
productive of any policy, or as enlightenment of the historical record. 42 Religious 
questions will be discussed below. 

More interesting is the suggestion of Dandamaev that Cambyses was disliked by 
the Persian aristocracy for his tendency to centralize power in his hands at the expense 
of the nobility, and that Gaumata/Bardiya sought the support of the people against 
the nobility which led to his overthrow and the "support of the nobility was 
for Darius." 43 The revolts against Darius then were 'popular' revolts against the 

40 It is not possible to list those for and against convinced many, otherwise he would not have 
the veracity of Darius in his inscription. Danda- been successful against many odds, and it would be 
maev, supra, Persien, 120; A. T. Olmstead, History difficult to prove that he lied. 

of the Persian Empire (Chicago, 1948), 93, and 42 Dandamaev, supra, Persien, 143-45, has con- 
others are contra. Hinz, op. cit. [n. 1], 132-33; S. vincingly downplayed these facets of the story, 
Shahbazi, Darius the Great (Shiraz, 1971); 7-11, though the anniversary of the killing of Gaumata 
Herzfeld, op. cit. [ch. 3, n. 31], 2, 205-09, and or Bardiya apparently was celebrated as somehow 
others support the veracity of Darius. connected with the Magi, and Dandamaev's 

41 E.g., the difficulty of uniting six noble dismissal (138-39) of this as a mere coincidence of 
Persians against a legitimate Achaemenid king, the killing of Gaumata/Bardiya with the festival of 
Bardiya brother of Cambyses. All arguments, bagayadi! is unconvincing. 

unfortunately, are subjective. Obviously Darius Ai Ibid., 157-67. 



100 Chapter V 

aristocracy headed by Darius, and Dandamaev translates an OP word vi&apatiy as 
simply 'house' in the Behistun inscription (II, 16 and III, 26), thus "the Median people, 
which were in (their) houses," and the "Persian people, which were in their houses," 
revolted against Darius, while the professional army and the aristocracy supported 
him. Note that this proposition is based primarily on the interpretation of one word, 
which is by no means universally accepted. 44 On the other hand the reasoning of 
Dandamaev is attractive; in effect it goes somewhat as follows: the centralized army 
of Cambyses returned to Iran to suppress a usurpation either by Bardiya or Gaumata, 
who was popular among most of the subjects of the empire, especially since a 
moratorium of three years on the payment of taxes had been declared (Herod., Ill, 
67). 45 Then Bardiya/Gaumata seized the property of the Persian nobility, who 
presumably were in the army of Cambyses, and destroyed the sanctuaries of these 
nobles, whatever they were. So Darius and the other Persian nobles, who supported 
him, killed and overthrew the usurper, and then restored the privileges and 
sanctuaries of the nobility. This explanation of the conflict between the central 
government (of Cambyses and the usurper) on the one side and Darius and the landed 
nobility on the other, does make a consistent sense, but it is again only a conjecture, 
but perhaps better than other surmises. This further implies that Darius was more the 
head of an oligarchy rather than an all-powerful potentate, and the correctness of 
Dandamaev's proposal can be measured only by the consequences or aftermath of the 
accession to the throne to which we must now turn. 

The rebels whom Darius had to suppress were mostly those who sought to 
reestablish their local independence, and only one Vahyazdata in Persis, by 
proclaiming he was Bardiya son of Cyrus, made pretensions to the Achaemenid 
throne. One was a local chieftain in Merv whom Darius, in order to increase his 
importance, raised to the rank of 'king' in Margiana (DB IV, 24). Two, Phraortes and 
Cissantakhma, claimed descent from the Median ruling house; further two, Nidintu- 
Bel and Arakha, each claimed to be the son of Nabonidus, last king of Babylon, while 
the Elamites were more rebellious than any other people mentioned in the 
inscription, their third rebellion occurring in 520 B.C. Several remarks come to mind; 
first, all the revolts take place in the heartland of the empire, on the plateau or the 
Mesopotamian plain. None of the revolts in the western possessions are mentioned in 
the Behistun inscription, although (II, 7) Egypt is recorded as a rebellious land. It may 
be that the inscription was intended primarily for the Iranians, who seem to have been 
the principal and most dangerous rebels, such that Darius' first year could be called 

44 The Akkadian version of the above passage, rebel, which implies a struggle for power rather 

not fully known to Dandamaev when he wrote, than a people vs. aristocracy struggle, 

says, "all the Persian troops who had previously 45 If Bardiya revolted against his brother, some 

come to me to the palace of Babylon from Anshan reason must have induced him to this action, and 

revolted from me and went over to Vahyazdata." furthermore the main army was with Cambyses. 

Cf. E. von Voigtlander, The Bisulun Inscription of Herzfeld, Zoroaster [ch. 3, n. 31], 1, 208, gives 

Darius the Great Babylonian Version (London, different motives for different revolts, including 

1978), 59. The text also could be translated "the the east Iranian revolts directed against the strict 

Persian troops as many as had been collected in the centralization of the Achaemenids. As noted, the 

house of Babylon from Anshan." This means that revolt of Vahyazdata in Persis, claiming to be 

the Persian troops who had come to join Darius at Bardiya, at least indicates great uncertainty in the 

Babylon instead returned to Persis and joined the minds of the people about Bardiya. 



Achaemenids 101 

one of civil war. Second, the rebels claimed legitimacy by family connections with 
dead rulers; whether they actually did have legitimacy or were liars according to 
Darius, they surely were all taking advantage of the end of the family of Cyrus and the 
usurpation of Darius (in their eyes). So, whether Bardiya was really himself or 
Gaumata using his name, many people thought a legitimate ruler had been killed by 
Darius, a usurper, and the Behistun inscription is Darius' greatest propaganda device 
against this position. The civil war then appears to have been more a struggle for local 
autonomy or personal competition with Darius for power rather than a series of 
battles embodying social classes or interests, as Dandamaev perceives. It was a time of 
confusion and one can well imagine that Persian troops on the way from Fars to 
Babylon, when they heard that someone calling himself Bardiya was raising a force in 
the homeland, would return to their homes to support one known for his remission of 
taxes and his benevolence, rather than to support a relatively unknown Darius. 46 
Again, there are so many possibilities of interpretation of motives that it becomes an 
exercise in futility to seek to divine the true intentions of the actors in the drama. 

Two other passages in the inscription relevant to the motivation for revolt, 
however, have been much discussed. After the overthrow of Bardiya/Gaumata, 
Darius says he rebuilt destroyed sanctuaries and returned to the people the fields, 
herds, slaves and houses which had been taken from them. 47 Why would Bardiya/ 
Gaumata destroy sanctuaries and confiscate property, and whose property? Since 
Darius restored everything as it had previously existed, it would seem that Bardiya/ 
Gaumata indeed had sought to change the status quo, which was then restored by 
Darius. Were the sanctuaries those dedicated to many minor or local deities, whom 
Bardiya/Gaumata sought to replace by a more restricted devotion to one deity? If 
Bardiya/Gaumata revolted from Cambyses, the revolt was also against the army 
generals with Cambyses, who had received fief lands from Cambyses. 48 The easiest 
explanation of the actions of Bardiya/Gaumata is that he gathered supporters and 
confiscated the property of the supporters of Cambyses, most of whom would be 
military officers. Then he destroyed their sanctuaries, but this action is unclear. Were 
many of the officers devotees of a cult to Mithra or to a Mdnnerbund fraternity? Was 
this action primarily religiously or politically motivated? We do not know, but we 
may believe Darius when he claims to have restored everything as it was previously; 
but was the second passage of the inscription a return or something new, the elevation 
of his accomplices to a special standing? 

The privileges of the Persian nobility existed before Darius, but it is possible that 
the seven great families of Iran, which we often meet in later sources, began with 

46 As a matter of methodology, one should von Voigtlander, op. cit. [n. 44], 55, has "I gave 

accept any additional information in one of the back to the army the herds, the flocks, the fields, 

linguistic versions of the Behistun inscription, (and) the hired workers, (comprising) the 'bow' 

since it adds information which easily could have estates which that Gaumata, the Magian, had taken 

been omitted from the other two languages since it from them." The qaiti 'bow' estates were fief lands 

was considered insignificant. Thus, the fact that the given to military officers in lieu of service. 

Persian troops from Anshan (Fars) were on the 48 Attested in Babylonia already in the first year 

way to Babylon, only in the Akkadian version, of Cambyses, 529 B.C. Cf. Dandamaev, "Die 

would likely only interest the Babylonians; Lehnsbeziehungen in Babylonien unter den ersten 

therefore it should be believed since it only adds Achameniden," Festschrift fur Wilhelm Eilers,ed. by 

and does not contradict. G. Wiessner (Wiesbaden, 1967), 36. 
The Akkadian version, in the translation of 



102 Chapter V 

him. 49 His six helpers are listed in the inscription, in Herodotus and by Ctesias. In 
Behistun IV, 80-88 we have: 

O.P. Vindaiarna son of VayaspaTa Akk. Vintaparna son of Visparu 

Utana son of Thukhra Vittana son of Sukhra 

Gaubaruva son of Marduniya Gubaru son of Marduniya 

Vidarna son of Bagabigna Vidarna son of Bagabigna 

Bagabukhsa son of Datuvahya Bagabukshu son of ZatQa 

Ardumanis son of Vahuaka Ardimanis son of Vakhku 

The Elamite version is either illegible or it copies the OP. Herodotus (III, 70) has 
'IvTaxplpvqs, OrdvrjS, rwfipv-qs, 'Y86pvr)s, Meyafiv^os and ^Aoirad-ivrjS, while 
Ctesias (14) has ^AraxpdpvTjs, , Ov6<pas, Map86vios, ^Ihipvqs, No(p)ov8afia.Tr}s, 
Bapiaar/s, and he adds two more later conspirators 'Apraovpas and Bay(maTr)s. S0 
We may take the Behistun inscription as the most reliable, but the substitution of the 
name Aspathines for OP Ardumanis is noteworthy, since Aspacina is depicted and 
named as the bowbearer, but holding a battle-axe beside the tomb of Darius at Naqsh- 
e Rustam. We may suppose that Aspacina was not one of the conspirators but later 
became powerful, and when Intaphernes was executed and his family fell from 
honor (Herod, III, 118-19), then Aspacina either took his place, or Herodotus 
thought he did. In any case, we hear of these families many times, that they owned 
land in various parts of the empire, that they held special privileges in court and the 
like. They may have provided the basis for the 'seven great families' of Iran in 
Parthian and Sasanian times with origins in Achaemenid times. 

It is not possible to discuss details of the Behistun inscription such as the 
chronology, for Darius gives month dates, and when he says he did so much in one 
and the same year (IV, 42-43), this has brought forth much controversy. 5 1 The entire 
veracity of Darius has been based on this assertion, but many are the possible 
explanations, and Darius would hardly put on stone and parchment, and send 
throughout the empire statements which obviously could be proved false, such as the 
adding up of dates. Whatever the dispute about the truth of his statements in the 
inscription, Darius won, and then he did so much it is difficult to even list his 
accomplishments. 52 Here we shall briefly speak of his further conquests, his 
reorganization of the empire centrally and the satrapal system, and finally religious 
questions and his buildings at Persepolis. 

49 The seven great Persian families became a SHAW, 3 Abh. (1977) 5-40, summarizes the 

tradition well known to outsiders. Cf Aeschylus, sources but does not add anything. 

956-60; in the Bible, Esther 1, 14, Ezra VII, 14, 51 E.g., from A. Poebel, "The Reign of Smerdis 

Josephus, Jewish Antiquities, XI, 3, 1, and others. and Others," AJSL, 56 (1939), 121-45, to A. S. 

so On the reading and etymologies of these Shahbazi, The 'one year' of Darius re-examined," 

names see R. Schmitt, "Medisches und persisches BSOAS.iS (1972), 609-14. Darius obviously was 

Sprachgut bei Herodot," ZDMG, 117 (1967), in a hurry to carve his inscription, since the 

120-22; Kent, op. cit. [n. 39], in the vocabulary, as Scythian campaign was added later in the fifth 

well as Brandenstein und Mayrhofer, Handbuch, column of the inscription, and the Egyptian 

and especially R. Schmitt, "Nachlese zur achai- reconquest is not even mentioned, 

menidischen Anthroponomastik," BZN, 6 (1971), 52 Attempts to prove that the insciption was 

1 -1 6. He argues for the reading Ardimanis for the written in verse, or to analyze all OP inscriptions as 

Artasyras of Ctesias. The study by F. Gschrutzler, following some scheme of organization are inter- 

"Die sieben Perser und das Konigtum des Dareios," esting, but do not throw light on the history of the 



Achaemenids 103 

FURTHER CONQUESTS 

After the civil wars were over, Darius in 519 B.C. says in the fifth column (20-30) to 
the Behistun inscription, which was clearly added later than the first four columns, 
that after suppressing the third Elamite revolt, "I went with an army against the 
Scythians (Saka). Then the Scythians who wear the pointed cap, these Scythians went 
away from me. (When) I came to the draya ('sea' or 'river'), I crossed it with all my 
army. Then I smote the Scythians mightily and another I took prisoner; this one was 
led bound to me and I slew him. The chief of them, by name Sku(n)kha, him they 
seized and led to me. Then I made another their leader, as was my desire." 53 One new 
theory proposes that this passage refers to the famous expedition of Darius against the 
Scythians of south Russia which then took place in 519 (Herod. IV, 1) and not in 513 
B.C. as usually supposed. 54 Without going into an exegesis of the fascinating 
information in Herodotus about that campaign and the description of the Scythians, 
suffice it to say that the Persians were not successful in holding these Scythians as 
subjects, but still a problem of identification of those Scythians remains. For Darius 
tells us (II, 8) that Egypt and Scythia (Saka) were among the lands which revolted 
against him after he had suppressed the first revolt in Babylon in 519 B.C. This implies 
that the Sakas had been part of the Achaemenid Empire under Cambyses, and 
probably under Cyrus, but which Sakas had Cyrus subdued - those in south Russia? 
Or did Darius lead an expedition against the Scythians of south Russia, supposing 
them to be the same people who had submitted to Cyrus in Central Asia ? Also Darius' 
inscription at Persepolis (DPe) includes Hindu! (India) in the lands of the empire but 
not Thrace (Skudra), which, however, does appear in the inscription on his tomb at 
Naqsh-e Rustam, presumably engraved late in his reign. This suggests that the 
campaign against Skunkha mentioned in the fifth column of Behistun was not the 
expedition against the Scythians of south Russia, rather it was an earlier campaign 
directed against the Scythians of Central Asia. 55 

times. See C. Hauri, Das pentathematische Schema (Saka) Campaign," Al, 4, Monumentum H. S. 

der Altpersischen Inschriften (Wiesbaden, 1973), and Nyberg 1 (1975), 77-88. See also the OP 

CI. Herrenschmidt, "Designation de l'empire et inscription on a clay brick from Gherla, Rumania, 

concepts politiques de Darius 1 d'apres ses inscrip- which might have some relevance to the presence 

tions en vieux-perse," SI, 6 (1976), 33-65. Her of Achaemenids in the Balkans, J. Harmatta, "A 

assertion that the word burnt means 'empire' for Recently Discovered Old Persian Inscription," 

Darius is hardly acceptable. AAH, 2 (1954), 1-14. 

53 (V, 21-30). There is no Akkadian or Elamite 5S It is not possible here to discuss the various 

version of column V, and the OP text is poorly Scythians (Saka) mentioned in the inscriptions, or 

preserved, so the translation is based on a number seek to locate them precisely. The Naqsh-e Rustam 

of hypothetical restorations. The section beginning inscription of Darius distinguishes between Scyth- 

"another I took prisoner" is unclear, but the lans with pointed hats, Amyrgian Scythians 

meaning seems to be that the Persians caught one (haumavarga) and Scythians across the sea (tayaiy 

Scythian chief and executed him, whereas paradraya). In spite of many other theories, the first 

Sku(n)kha, the primary ( ?) chief was captured but two Scythians are to be located in Central Asia, and 

not killed. Also many scholars have taken the word the last probably in the Balkans or south Russia. 

draya as meaning 'river' and have put the action in The Egyptian expressions on the Suez stele and the 

Central Asia. Others claim the river was the Susa statue, 'Sakas of the marshes' and 'Sakas of the 

Danube. plains,' seem to be an Egyptian manner of 

5i Cf. J. M. Baker, "The Date of Herodotus IV, expression and do not aid us in locating them. Cf. 

1 , Darius Scythian Expedition," Harvard Studies in Kervran, "Une Statue de Darius decouverte a 

Classical Philology, 76 (1972), 99-132, and G. C. Suse," JA (1972), 258. 
Cameron, "Darius the Great and his Scythian 



104 Chapter V 

Egypt was brought back into the empire probably in the winter of 519— 518 B.C. by 
Darius himself, seemingly after he had issued a decree permitting the Jews to finish 
building the temple at Jerusalem (Ezra VI, 1). Probably at the end of 518 B.C., after he 
had left Egypt, Darius gave orders for the codification of Egyptian laws, which was 
followed elsewhere such that Darius was called a great lawgiver. 56 To return to his 
conquests, the Indian campaign meant a great extension of the empire. 

The date of the Indian campaign can be determined by a comparison of the 
Behistun inscription where Hindus is not mentioned as one of the countries which 
'came to' Darius, whereas in Persepolis (DPe) and in Naqsh-e Rustam Hindus is added 
to the list of countries. Therefore, in the first part of his reign Hindus was at some time 
added to the other eastern possession, Gandhara (where the Akkadian version of 
Behistun 1, 1 6 has pa-ar-u-pa-ra-e-sa-an-na or Avestan upairisalna 'above the eagle") and 
Arachosia. As Herzfeld noted, since the voyage of Skylax, the Carian admiral of 
Darius who was sent down the Indus River and back to the Red Sea on a voyage of 
exploration, probably took place after the conquest, because a 'royal garrison' is 
mentioned in a fragment of Hekataios' account of the journey as preserved in Stephan 
of Byzantium, 175, the date of the conquest would be between 518 and 515 B.C. 57 The 
limits of Hindus, however, are by no means clear and most scholars have considered it 
only the Indus valley south of Multan to the sea. More likely, however, the name 
came from the river, the Sindhu, and would include the people living in the entire 
valley including the Punjab, and the border with Gandhara would be the lowlands of 
the present NW frontier province extending to the mountains. 58 The voyage of 
Skylax opened sea communications between India and Egypt and probably induced 
Darius to build a canal in Egypt. The idea for this canal was an incomplete project 
started by a pharaoh Necho (c. 610-595 B.C.), but Darius finished a canal from the 
Nile to the Red Sea and erected a number of steles in commemoration of the opening 
of the canal, probably in the year 498 B.C. 59 Statues of Darius were also erected in 
various temples in Egypt and Darius took his place as a legitimate pharaoh of Egypt in 
the eyes of the people. 

The last expedition, recorded by the sources, is the famous invasion of Greece 
which ended at Marathon in defeat presumably at the end of August 490 B.C., 
although the exact date is uncertain. It was the aftermath of the long Ionian revolt 
on the coasts and islands of Asia Minor, which lasted c. 499 to 493 B.C. and was 
suppressed only after much fighting by the Persians. Some scholars have regarded the 
Ionian revolt and the battle of Marathon as insignificant in the eyes of the great king, 
proposing this as a kind of balance to the obviously biased Greek sources, but the 

56 For the Egyptian sources see R. A. Parker, 5S Many scholars consider that the entire Punjab 
"Darius and his Egyptian Campaign," AJSL, 58 was included in Gandhara, e.g., Breloer, op. cit. [n. 
(1'941), 373-77, esp. note 1. The Persian governor 57], 27, but it is more likely that Gandhara only 
of Egypt Aryandes was executed by Darius ace. to extended to the modern boundaries of the Punjab, 
Herod. (IV, 1 66), who gives several reasons for this but including Taxila. Cf. Tarn, op. cit. [ch. 1 , n. 23], 
action. 135, and V. Agrawala, India as known to Panini 

57 Herzfeld, op. cit. 2 [ch. 3, n. 31], 662, and a (Lucknow, 1953). 50. 

discussion in detail with the same remark in B. 59 M. Kervran, op. cit. [n. 55], 266; and W. 

Breloer, Alexanders Bund mil Poros (Leipzig, 1 941 ), Hinz, "Darius und der Suezkanal," AMI, 8 (1 975), 

5-17. Presumably by a 'royal garrison' Hekataios 1 15-21. 
meant Achaemenid, although this is not certain. 



Achaemenids 105 

pendulum should not swing to extremes. The Achaemenids found Greeks in Egypt 
and in Cyprus, and cannot have regarded them as insignificant, since the Greeks 
dominated the seas perhaps even more than did the Phoenicians, who were more 
loyal subjects of the Persians. Furthermore, the cultural and technical achievements of 
the Ionians were also known and appreciated by the Persians as testified by the many 
Ionians in trusted positions under the Achaemenids. 60 The Ionian revolt, which it 
must be remembered included Greeks in Cyprus and Carians on the Anatolian 
mainland, as well as the Ionians, culminated in the capture and burning of Sardis in 
498 B.C.; and this must have persuaded the Achaemenids that the Greeks were much 
more than a mere thorn in their far western side. Much has been written about the 
Persian support of the Greek tyrants in the Ionian cities, who were overthrown and 
driven out by the people, as well as economic reasons for the revolt in Ionia - fears 
that the Persians would manage trade by their control of the Bosphorus and the grain 
trade of south Russia, as well as Egyptian trade. 61 It was inevitable that the mainland 
Greeks would also become involved thus leading to Marathon. But before the 
expedition to Greece, Thrace and Macedonia had to be reconquered after the Ionian 
revolt, an undertaking headed by Mardonius, son-in-law of Darius which cost many 
lives and hardships (Herod. VI, 44). 62 Thus the expedition to Greece headed by Datis, 
a Mede, was not simply an expedition of revenge on mainland Greece for support of 
the Ionian revolt, but was a continuation of Persian policy to secure a position as 
master of the seas, while the Greek building of a fleet and reconquest of some of the 
islands of the Aegean Sea after Marathon indicates that the conflict was perceived as 
serious and continuous by both sides. 

Two warnings about the Classical sources should be kept in mind. First, the 
tendency to assign personal motives for all events is dominant in Herodotus and other 
authors, and the stories told to explain why something happened, although 
entertaining, and in many cases even important for our understanding, nonetheless 
must be regarded with great circumspection. Names of generals, governors and 
others abound as actors in a drama, but in a survey of the history of Iran even accounts 
of those important Iranians, such as harem figures and governors of provinces, cannot 
be elaborated or even discussed. Second, the tendency to use fixed figures and standard 
plots in describing events does not persuade one to have confidence in the information 
given. For example, the number six hundred is the standard for ships in the 
Achaemenid fleet in Herodotus (IV, 87 on the Scythian campaign, VI, 9 by Lade and 
VI, 95 for Marathon), while numbers like 700,000 soldiers on the Scythian campaign 
(IV, 87) are absurd. That the Greeks were better armed and better protected with 
armor than the majority of the Achaemenid army is generally accepted, and as time 

60 See A. R. Burns, Persia and the Greeks 62 The expedition to the north has been 
(London, 1962), 241, and n. 10. considered a first attack on Greece which failed 

61 See J. M. Cook, The Greeks in Ionia and the because of storms offMt. Athos which destroyed 
East (London, 1962), 98-120, 132-33; and C. much of the Achaemenid fleet, and because of 
Nylander, Ionians in Pasargadae (Uppsala, 1970). fierce resistance in Macedonia. More likely was the 
The 'royal road' from Susa to Sardis indicates the need to first secure the northern areas which had 
importance of western Asia Minor and the islands broken away from Persian rule before dealing 
>n the minds of the Achaemenids. See discussion directly with mainland Greece. 

and bibliography in H. Bengtson, op. cit. [n. 18], 
156—61. 



106 Chapter V 

progressed the Greek mercenaries became the professional core fighting force of the 
Achaemenid army, while Persians lost both their ability and their will to fight. 63 But 
in Darius' time they were still good soldiers, though poor sailors. Darius, even in his 
old age, was determined to counter-attack Greece by land, and began preparations for 
a long campaign, but Egypt rebelled in 486 and in the same year in November Darius 
died, according to Akkadian documents. 64 



REORGANIZATION OF THE EMPIRE 

Darius seems to have been the first of the Achaemenid great kings to take a 'throne- 
name' since his meant 'possessing good(-ness),' and his successors also selected throne- 
names. 65 Whether his own name was Spentadata, which Ctesias (frg. 10) claims was 
the real name of Gaumata, is impossible to determine. 66 If we consider the etymology 
of Xerxes (OP KMayarsan) as 'ruling over heroes' and Artaxerxes (OP Artakhssasa) 
'having just rule,' the religious nature of them is apparent, but it is interesting that only 
the names Darius and Artaxerxes were taken by the later kings of the dynasty. 67 The 
names were a proclamation of the right to rule, but by their own qualities, as well as 
by divine intervention, as found in the inscriptions, where we see that it was by the 
will of Ahura Mazda that rule was bestowed on the kings. The king was the supreme 
arbiter, above all men, and his exalted position was proclaimed by court ceremonies, 
especially by proskynesis, which meant kissing the king's feet for servants, to a cheek 
kiss for the nobility. 68 The title 'kings of kings,' as already noted, was probably used 
by the Medes, although only its phonetic form suggests this, since no inscriptions of 
the Medes have been found. 69 The crown prince was obviously next in court, but 
some scholars have confidently postulated the existence of an institution of double 
kingship under the Achaemenids. 70 Why one should confuse the 'institution' of 

63 By far the greatest number of Greeks serving feld, op. cit., 1 [ch. 3, n. 31], 97, 'who commands the 
the Achaemenids named in the sources were in right will,' and Kent, op. cit. [n. 39], 182, 'hero 
military service. Cf. J. Hofstetter, Die Criechen in among kings' (unlikely). The name 'Xerxes' was 
Persien (Berlin, 1978), 216 pp. only used once. Cf. K. Hoffmann, "Altpers. 

64 R. A. Parker and W. Dubberstein, Babylonian afuvaya" Corolla Linguistica, Festschrift F. Sommer 
Chronology (Providence, 1956), 17. (Wiesbaden, 1955), 85, n. 15. 

65 It is interesting that in Aramaic documents, 68 Cf. J. Hofstetter, "Zu den griechischen 
dated c. 480 B.C., the name is drwS, later became Gesandschaften nach Persien," Historia, Einzelschr- 
drywS (also in the Bible), and for Darius II we have iften 18 (Wiesbaden, 1972), 106, with further 
dryhwS or drywhwS, thus the later the time the bibliography. The kings were not divinised, as 
longer the name, contrary to what one would some have thought. On proskynesis see F. Altheim, 
expect. On throne-names see R. Schmitt, "Thron- GesMchte der Hunnen, 2 (Berlin, 1960), 125-52. 
namen bei den Achaimeniden," BZN, 4 (1977), 69 Szemerenyi, op. cit. [ch. 4, n. 63], 321, claims 
422-25, who convincingly suggests that Darius that the Medes did not use the title since they were 
innovated the practice of throne-names as part of under the suzerainty of the Assyrians and could not 
his overall policy to raise the place of kingship in use such a proud title. After 612 B.C., in any case, 
the eyes of his subjects. Herzfeld, op. cit., 1 [ch. 3, n. this statement is nonsense, if not previously as well. 
31], 1, 89-99, first proposed the idea of 'throne Although the title Sar-sarrani ''king of kings' is well 
names.' attested among the Old Assyrian kings, we do not 

66 Herzfeld, op. cit., 1 [ch. 3, n. 31 ], 95, identifies find it attested for the neo-Babylonian kings. Cf. 
Darius with the son of the Vishtaspa who was a M.-J. Seux, Epithets royales akkadiennes el sumer- 
patron of Zoroaster, the IsfandiySr of Firdosi, iennes (Paris, 1967), 318. 

which is unlikely. 70 Ibid., 365, and from the artistic side see H. von 

67 On other etymologies for Xerxes, see Herz- Gall, "Die Kopfbedeckung des persischen Ornats," 



Achaemenids 107 

crown-prince with the idea of a dual kingship I do not understand, for neither a 
crown-prince nor an 'advisor' to the king, or chief minister, or whatever he may be 
called, can be designated a 'double king' as we find among later Turkish kingdoms of 
Central Asia, or in ancient Sparta. If there was no institution of a 'double king' in 
ancient Iran, then did the crown prince have a special title? In Akkadian we have the 
term for 'crown prince* mar iarri used of Asarhaddon and Assurbanipal, but whether 
this was also used of other sons of the king as well as (or only) the crown prince is 
unknown. 71 Many scholars have mistakenly assumed that the co-regency of king and 
crown prince at the end of the king's reign is proof of the institution of 'double 
kingship,' but this practice of associating one's successor in rule as a kind of apprentice 
is found throughout the history of Iran, and elsewhere in the world, and should not be 
confused with an institution of 'double kingship.' 72 Likewise the office of 'second to 
the king,' a major-domo or top minister of court, is normal throughout history, and 
again does not mean co-king. Was there a 'second' or assistant to the king under the 
Achaemenids? Again no title is attested, but in later Parthian and Sasanian times, we 
do find a title bidakhsh which may go back to an Old Iranian *dvitiya-khfaya - 'ruling 
as second,' with an extensive literature about it. 73 It is uncertain, however, whether 
the notices about a 'second' after the king refer to a title in the governmental 
organization or in the court, or to an honorary appellation given to a favorite of the 
ruler. In either case it does not refer to the crown-prince, who was surely the real 
'second' after the ruler, and there is no title attested in the sources. Greek authors, 
however, imply that a title ^tAiap^os, 'a thousand leader,' meant 'second* (after the 
king), although this is probably by origin a military title, the equivalent of Iranian 
*hazdrapati-, attested in languages other than Old Persian. From researches of 
Marquart and Junge it seems that the chiliarch, as commander of the royal guard and 
probably also of the 'immortals' (royal guard), did act as an usher of persons who 
wished to speak with the ruler and consequently could be regarded as a kind of 
minister of court or prime minister by the end of the Achaemenid dynasty. 74 Thus 
the power of the chiliarch grew as time progressed with both executive orders and 

AMI, 7 (1974), 157-59 and especially P. Cal- (Vienna, 1938), 114-15. Akkadian documents 

meyer, "Synarchie," AMI, 9 (1977), 63-68, who confirm the association of the king's son as 

writes, "Das persische Doppelkonigtum ist also von successor, but never as co-king with the same titles 

Dareios I/Xerxes I bis zu Artaxerxes II/Dareios as his father. When Xerxes says (XPf) that Darius 

belegt." And not afterwards? had other sons but made him 'greatest' maBiSta after 

71 Seux, op. cit. [n. 69], 160. It was also used in himself, this is hardly a title. In India the crown 
late Akkadian documents for Cambyses and in the prince was called yuvarllja 'young king,' but again 
time of Darius II, so it continued in use, but, of no 'double' kingship. 

course, we still do not know whether it was used 73 W. Hinz, Altiranische Funde und Forschungen 

exclusively for the crown prince or for other (Berlin, 1969), 153, note 22, where further 

princes as well. For the designation of Cambyses as bibliography may be found. The expression 'after 

'king's son' as recognition that he was his father's self also implying second authority, is a later 

successor, see W. Dubberstein, The Chronology development. 

of Cyrus and Cambyses," AJSL, 55 (1938), 417- 74 For references see J. Marquart, Untersuchun- 

19 - gen zur Ceschichte von Eran (Gottingen, 1896), 57- 

72 The co-regency of Cambyses with Cyrus, 63; E. Benveniste, Titres et noms propres in Iranien 
before the latter left on his eastern campaign which ancien (Paris, 1966), 67-70; and J. Junge, "Zur 
led to his death, has been taken wrongly by many Stellung des Chiliarchen der kgl. Leibgarde im 
as proof of the existence of an institution of 'double Acha'menidenstaat," Klio, 33 (1940), 13-38. 
kingship'; cf. F. W. Konig, Der Falsche Bardija 



108 Chapter V 

control of finance under him, as well as control over the court and protection of the 
person of the ruler. If we may assume, as Junge argues, that the chiliarch did become 
'second' after the king, then we may conclude that any expressions denoting 'second' 
were not official titles in the government but rather appellations, either bestowed by 
the ruler or current among the people as a matter of designation of the great power of 
an official, second only to the ruler. 75 

We cannot clearly discern whether there was a bipartite or even a tripartite 
division of authority in the Achaemenid Empire, as appeared later in the history of 
Iran. The three parties wielding power were the court, the bureaucracy (perhaps 
including priests) and the army. Inasmuch as the chiliarch combined military and 
administrative functions, as did the satraps, we may assume such a mixture in other 
high offices. At the court of the great king, several positions of influence stand out 
from the time of Darius. One was the 'bow carrier' uassabara and the other the 'spear 
bearer' arstibara of Darius, both 'honorary' appellatives given to close friends of the 
ruler, who might be considered as his bodyguards and confidants. 76 At the court 
royal princes, of course, had great influence simply as members of the royal family 
(Aramaic BR BYT, O. Iran. *visapu\}ra) and other nobles (amata), both Iranian and 
non-Iranian, gave advice to the king. We do not know whether a council of advisors, 
perhaps the six helpers of Darius in his time, existed as a kind of cabinet under later 
Achaemenids, for there is again no evidence. Perhaps there was an informal, indefinite 
group of advisors, the 'drgzry' or *handarzakara?. of Daniel III, 3, who may have been 
part of a hierarchy of friends around the king. Greek sources have much information 
about confidants of the king but the Greek words may apply to Greek interpretations 
rather than Persian realities, especially when the Persian words for the Greek terms do 
not appear anywhere. The Greeks, for example, speak of institutions called 'the king's 
friends' (piXoi and 'the king's benefactor' £vtpyirt)s, these latter according to 
Herodotus (VIII, 85) called in Persian opoadyyai. 11 There may have been divisions 
in the category of 'friends,' some greater (Diodorus XV, 10, 3), others lesser, as well as 
different degrees of closeness to the king among the kinsmen of the ruler. The 
eunuchs, harem ladies, princesses (duxsi), made of the court a large collection of people 
using their influence on the ruler. 78 Also probably connected directly with the court 
were the famous 'eyes and ears' of the king (Xenophon, Cyrop. VIII, 2, 10-12) which 
have produced a large literature. 79 It seems we are dealing with a chief overseer, the 
'eye,' and many spies of the king, the 'ears,' who reported to him about affairs in the 

75 W. Hinz, "Achamenidische Hofverwaltung," ZDMC, 117 (1967), 131. The identification of 

ZA, 61 (1971), 303 proposes a title *vi$a-palis (piXos in Iranian as Gathic rvaBa- by Herzfeld, op. 

'Hofmarschall,' for the second in authority, as a cit., 1 |ch. 3, n. 31], 155-56, is unconvincing. More 

continuation of Akkadian rabekalli 'palace chief.' likely is dauilar, or another word. 

His theory in Altiranische Funde und Forschungen 78 Cf. Benveniste, op. cit. [n. 74], 49-50. One 

[n. 73], 63-68, that this officer was always a Mede should not forget the education of young nobles 

is unacceptable. It is possible that the chiliarch later and princes at the court (Xenophon, Cyrop. VIII, 6, 

took over the duties of an earlier majordomo, but 10 and Plato, Alcibiades, 121-22) since the strict 

again there is no evidence. Persian rules of raising the young impressed the 

Hinz, Wege, [ch 4, n 66], 57-59, interprets Greeks. Unfortunately, sources are lacking on 

the first as 'clothes carrier' who carries the bow and details of the education, 

arrow for Darius at DNd. 7 » On the ocpdaX/xos [SaoiXevs see Herodotus I, 

For an etymology of this word possibly 114, Plutarch, Artaxerxes 12, 1, Aeschylus, 979, 

meaning 'widely known,' see R. Schmitt, "Me- etc. Among the possible Iranian words proposed 

disches und persisches Sprachgut bei Herodot," for the 'eye,' are the following: *spasaka, ctxia, 



Achaemenids 109 

empire, especially in the provinces, independently of the governors and military 
commanders, in other words a secret service of spies reporting directly to the court. 
Every official, courtier, indeed theoretically everyone, was a bandaka or 'slave' of 
the king, but the word was used probably more as a form of address than anything 
else. 80 There is another word anusiya 'follower,' which seems to have been mis- 
understood by the Greeks who translated it as dtMvaro? 'immortal,' for the 
royal guard, which was theoretically maintained at the same size at all times. 81 

Information was brought to the king by the famous postal system, which was not 
new but was reorganized by Darius, with the main artery of communications the 
'royal road' from Sardis to Susa (Herod. V, 52). Other roads also existed, especially a 
continuation of the 'royal road' from Susa to Persepolis and Pasargadae, traces of 
which have been found by archaeologists. 82 According to Herodotus these roads 
were measured in parasangs (c. 6 km.), and he describes the stations and system of 
couriers on the royal road (VIII, 98). 83 The Achaemenids were also active in the 
construction of bridges over rivers, in the use of boats for commerce and 
communications, as well as fire signals for the latter. 84 Although all of this activity 
was not solely for the benefit of the king, his orders played a dominant role in the 
creation of the system of transportation and communication. 

The Elamite tablets from Persepolis have not only given an enormous increase in 
Old Iranian onomastica but also new information about the chancellory, the 
organization of work, and religious practices in the homeland of the Achaemenids. 
The chief of court or major domo at Persepolis under Darius was a certain Farnaka until 
497 B.C., followed by Abisvanta to 484 B.C., then Dargayus to 482, Rtatakhma to 466 
followed by RtasQra. 85 The picture of a complex bureaucracy with many officials is 
clear from the tablets, but it is not possible to put them in a hierarchy or order. The 

*kasaka, *didlka or *dltaka, *patyaxla-\ cf. A. pecially in regard to the mythological and 
Pagliaro, "Riflessi di etimologie iraniche," Rendi- religious ramifications. 
conli delta classe di scienze morali, Accademia 81 Cf. Pagliaro, op. cit. [n. 79], 151. 
Nazionale dei Lincei, Serie 8, vol. 9 (Rome, 1954), 82 Stronach, Pasargadae [supra, n. 1], 166-67, 
133-53, with references; also H. Lommel, "Die and M. T. Mostafavi, "The Achaemenid Royal 
Spaher Varuna und Mitra," Oriens, 6 (1953), 323, Road," A Survey cf Persian Art, 14 (Oxford, 1967), 
who proposed *spadaka. Further W. Hinz, Ahira- 3008-10, who traces the road through modern 
niscltes Sprachgut der Nebeniibertieferungen (Wies- Firuzabad, ignoring the Anshan-Pasargadae part of 
baden, 1975), 105. Another title in the Elamite the road, which is far to the north and thus con- 
tablets from Persepolis has been reconstructed as tradicts the southern path proposed by Mostafavi. 
*apiy-axSa-pa 'controller of the overseers'! by M. 83 On the etymology of parasang see H. H. 
Mayrhofer,"DieNebenUberlieferungdes Altwest- Schaeder, Ungarische Jahrbiicher, 15 (Berlin, 1935), 
iranischen," ArchivfOr Orientforschung, 25 (1974- 563, and for the word ayyapos 'postal courier' see 
77), 182. The 'eye' may well have been a 'chief R. Schmitt in Glotla, 49 (1971), 97-100. 
inspector' for the king. 84 Ace. to S. Mazzarino, "Le vie di communica- 
The ears una or gaulaka are found in an zione fra impero Achemenide e mondo Greco, La 
Aramaic papyrus as gw!ky\ A. Cowley, Aramaic Persia e il mondo Greco-Romano, Ace. Naz. dei 
Papyri of the Fifth Century B.C. (Oxford, 1923), Lincei (Rome, 1966), 75-83; the Greeks learned 
99-102, and may be spies. Cf. H. H. Schaeder, much from the Persians in matters of 
Iranica, AGWG, 1 (1 934), 3-24. It is also possible communication. 

that these terms are Greek misunderstandings of 85 See Hinz in ZA, 61 (1971), 308. There were 

offices which did not exist. many scribes and assistants busy in the chancellory 

For a discussion of bandaka see Geo Widen- with a complex procedure of dictating, engraving 

gren,"Lesymbolismedelaceinture,"/.4,8(1968), and registering clay tablets in Akkadian and 

133-46. The remarks on marlka 'member of a Elamite, as well as scribes writing Aramaic on 

Ma'nnerbund' should be read with caution, es- parchment. 



110 Chapter V 

divan (Olr. *dipi-pana) or chancellory of later times surely goes back to an 
Achaemenid prototype, and in Persepolis many titles have been found for persons in 
the bureaucracy, the treasurer *ganzabara (attested in Aramaic and Elamite), his 
assistant the upaganzabara, then the jramatar and theframana-kara both in charge of, or 
directors of, work, the baji-kara 'tribute collector ,' the *hamara-kara 'tax collector,' the 
*gfdapati 'foreman' over workers, the raucapana 'overseer of daily rations,' and many 
other officials. 86 It is difficult to separate from each other the functions of many of 
these officials and to determine who controlled what, for example, who could use the 
king's seal, for there were many copies of the seal used by various officials. Another 
question is how much the organization of officials at Persepolis was common to other 
sites such as Susa, or Babylon, and how much was duplicated in the provincial 
organization. On the whole, it seems that the provincial courts tried to copy the 
imperial court, but there were obviously differences in various parts of the empire, 
but we have no direct information about this and subjective impressions may give 
false ideas if used as proven fact. 

When we turn to the provincial organization, in spite of copying from a central 
model, the variety rather than uniformity of institutions is striking. Two questions 
which may be elucidated are the role of local dynasts in the provincial organization of 
the empire and the subdivisions of the satrapy. A few local rulers like the Syennesis of 
Cilicia, probably the personal name of the ruler in the time of Cyrus which then 
became a hereditary appellation, the ruler of Paphlagonia (Xenophon, Cyrop. VIII, 6, 
7-10) and others held special positions in the empire because of their aid to Cyrus, or 
otherwise. In the revolt of Cyrus the younger against his brother Artaxerxes II, 
however, the Syennesis lost his throne for support of the younger Cyrus and his place 
was taken by a Persian. 87 Many local princes such as those of the Phoenician cities, 
Tyre, Sidon, Byblos, etc., tyrants of the Ionian cities, as well as the kings of Caria, 
Cyprus and other areas in the north and east where we have no information, were on 
the whole hereditary or at least in some tributary or vassal relationship with the 
Persians. What Darius did was to create twenty great satrapies over many of the local 
rulers, according to Herodotus, but this was not an innovation, rather a reorganization 
of the multifaceted parts of the empire which had existed under Cyrus and Cambyses. 
The former had divided Lydia into two satrapies with capitals at Sardis and Daskylion, 
and other changes had occurred until Darius' reform of the satrapies which we find in 
most detail in Herodotus (III, 89-94). The word 'satrap' was frequently used in 
Classical sources both for the governor and sub-governor as well. 88 

86 See the books of Bowman, Cameron, Hal- See also the references in R. Schmitt's survey, 
lock as well as the bibliographies of articles about "Forschungsbericht" [n. 29]. The fluctuation of 
the Elamite and Aramaic sources in M. Mayrhofer, titles and offices is revealed in the Persepolis 
"Neuere Forschungen zum Altpersischen," in Elamite tablets; see R. Hallock, The Evidence of 
Douum Indogermanicum, Festgabe fur Anton the Persepolis Tablets," CHI, 2. 
Scherer (Heidelberg, 1971), 41-66. There were, of 87 Cf. W. Judeich, Kleinasiatische Studien (Mar- 
course, other officials as well as various designa- burg, 1892), 209, and the detailed article on 
tions of them. One official the *padpabaga 'Satrap' by F. C. Lehmann-Haupt in PW. 
provided provisions for the road for other officials, 88 On the etymology of the word 'protector of 
while various inspectors undoubtedly existed, but the kingdom,' see the detailed exposition of R. 
one must not confuse synonyms, or public as Schmitt, "Der Titel 'Satrap'," Studies in Greek, 
opposed to official usage, as does J. Harmatta, "The Italic, and Indo-European Linguistics offered to L. 
Rise of the Old Persian Empire," AAH, 19(1971). Palmer, ed. by A. M. Davies and W. Meid 



Achaemenids 111 

Much has been written about the lists of lands or people in the OP inscriptions, the 
reliefs of tribute bearers at Persepolis, and the names of satrapies in Classical sources, 
many authors seeking to reconcile the three. 89 It is clear that the inscriptions speak of 
lands or peoples of the empire rather than satrapies (the word 'satrapy' is not 
mentioned), whereas the tax-list of Herodotus is defective; but it is possible to secure a 
measure of agreement of the various sources for the satrapies of the empire from the 
time of Darius to the end of the empire. Persis (P3rsa) was the homeland, especially 
supplying soldiers and bureaucrats, and it extended to the ocean on the south and 
included Kerman in the east. Although uncertainty exists about the eastern borders of 
the satrapy, it would seem that thoughout most of the period it included most of 
modern Kerman, although Alexander found a satrap of Kerman which implies a 
separation of the eastern part of Persis before the fall of the Achaemenids. Pasargadae 
(*parsa-argada 'settlement of Persians') and Persepolis (OP PJrsa) were the two 
capitals, the older and the younger. 90 

Media (Mada) was not free from taxation but nonetheless held a special position, 
since the Medes and Persians were associated together in the minds of other peoples of 
the empire (e.g., Esther 1, 19, Daniel VI, 8, 15). The satrapy of Media extended to the 
upper Tigris in the west and in the east to the 'Caspian Gates,' located east of modern 
Tehran, to Persis to the south and Armenia to the north, for it seems that Darius 
instituted a satrapy of Armenia, separating it from older and 'greater' Media. The 
Armenians are mentioned in the various lists of peoples in OP inscriptions, and they 
are represented on reliefs at Persepolis, but the northern extent of the Armenian 
satrapy cannot be determined, for the Georgians, Albanians and other peoples of the 
Caucasus Mts. apparently were never integrated into the empire although some of the 
Caucasian tribes were according to Herodotus (III, 94). The peoples of the south 
Caspian Sea coast were at some time separated from Media and made into a satrapy, 
probably after Xerxes, but at all times central authority was difficult to maintain in 
this area. Another important satrapy was Elam or Khuzistan (OP Huza) with its 
capital Susa, probably absorbed by Cyrus after his conquest of Babylon, which 
maintained its old boundaries of lowlands and mountainous regions to the time of 
Alexander. 

The eastern Iranian provinces and India present problems of geographical 

(Innsbruck, 1976), 373-90. The Persian rather Cameron, "The Persian Satrapies and Related 

than Median form of the word is found in OP Matters," JNES, 32 (1973), 47-56. For the 

khshassa-pavan which implies an Achaemenid Classical sources see the detailed article on 'Satrap' 

origin rather than a Median origin of the word. In by Lehmann-Haupt in PW, and for a synthesis E. 

the OP inscriptions we find dahyu- 'land' and not Herzfeld, The Persian Empire (Wiesbaden, 1968), 

the word 'satrapy.' Darius' attempt to separate the 288-365, comparing lists of Hekataios, Herodotus, 

military from the civil functions of the satrap was the OP inscriptions, etc. G. Walser follows 

unsuccessful later, while the number of satrapies Herzfeld in his "Missing Peoples in the Persepolis 

increased as time went on. Diodorus (XVIII, 5-6 Procession," Proceedings of the Fifth International 

and 39) gives the satrapies at the end of the empire, Congress of Iranian Art and Archaeology, 1 (Tehran, 

but this list must be used with caution. 1972), 372. 

89 On the reliefs see G. Walser, Die Vdlkerschaf- 90 Of the many etymologies proposed for 

leu aufden Reliefs von Persepolis (Berlin, 1966), and Pasargadae, that of H. W. Bailey mentioned here is 

W. Hinz, "Die Volkerschaften der Persepolis- the most convincing; cf. Stronach, Pasargadae 

Reliefs," Altiranische Funde und Forschungen [n. 73], [supra, n. 1 ], 280-81 . 
9 5-l 13. On the OP inscriptions see G. C. 



112 Chapter V 

identification. Parthia, to the east of Media, included Hyrcania (OP Varkana) 
modern Gurgan ; thus it comprised the modern provinces of Khurasan and Gurgan 
with a capital at Zadrakarta near the southeastern shore of the Caspian (Arrian, III, 25) 
and possibly a second capital in the east, although Herzfeld's assertion that *Tausa 
(TOs) was that capital finds no echo in any Classical source. 91 The suggestion of 
Pyankov that Parthia was ruled by a hereditary dynast instead of an appointed ruler 
on the basis of Ctesias, is not at all convincing. 92 Hyrcania was probably made into a 
separate satrapy under Xerxes (Herod. VII, 62), possibly because of its importance as a 
border province against nomads, but at the end of the empire it and Parthia were 
reunited (Arrian III, 8, 4). The Gils and Cadusians may have been united with 
Hyrcania at times, but we do not know. 

The Khwarazmians are placed in the same satrapy as Parthia by Herodotus (III, 93), 
presumably reflecting the time of Darius, whereas later they are separate, and by the 
end of the empire they were independent. This may well reflect their movement 
from the region of the Murghab River north to the Aral Sea, and for a time they seem 
to have comprised a separate satrapy. The belief that the Oxus River once flowed into 
the Caspian instead of the Aral Sea is not attested for historical times by archaeologists, 
so the Kharazmians are not to be placed on the eastern shores of the Caspian. The 
Sogdians most of the time were joined to the satrapy of Bactria, as was also Margiane 
or Merv, in the time of Darius, although later Margiane was joined to Aria (Herat). 
The satrapal capital was at Bactra, modern Balkh, with another center at Marakanda 
(Samarqand). 93 The independence of local lords in Sogdiana is attested by the fighting 
which Alexander did there; unfortunately we cannot say when Sogdiana was a 
separate satrapy and when combined during the Achaemenid period. Aria (OP 
Haraiva), although included in the large east Iranian satrapy of Parthia and 
Khwarazm by Herodotus (III, 93), was thereafter made into a separate satrapy with its 
capital Artacoana (with many variants) probably near the site of modern Herat. 94 
Drangiane (OP Zranka) apparently was joined with Aria for tax purposes (Strabo X, 
516), although Herodotus (III, 93) puts the Zarangians (also with S-) together with 
inhabitants of the central deserts of Iran through Baluchistan to the ocean. At the end 
of the empire Drangiane was together with Arachosia under one satrap ace. to Arrian 
(III, 2, 1), while the name of the principal city is Greek Prophthasia in Strabo (514 and 



91 Herzfeld, op. fir., 318. His corrections of wards in time by Ctesias is to be regarded with 
Arrian and Ptolemy may be correct, but Tos was great suspicion. 

not an important center. Although some sources 9i The etymology of Samarqand proposed by 

imply that Hyrcania was detached from Parthia A. Pagliaro, "Cyrus et l'empire Perse," AJ, 2 

and attached to Media it is impossible to trace (1974), 6, zamar kanta 'dug in the earth' is 

changes in the boundaries of satrapies which lasted unconvincing since the latter part really meant 

short times. On Jews deported to Hyrcania see J. 'town,' to be sure from the root 'to dig,' but the 

Marquart, Fundamente israelitischer und judischer former part was a proper name or appellation 

Geschichte (Gottingen, 1896), 30. rather than a common noun. See also I. V. 

92 1. V. Pyankov, "Istoriya Persii Ktesiya T Pyankov, Drevnyi Samarkand v izvestiyakh antich- 
Sredneaziatskie Satrapi Akhemenidov," VDI, 2 ttikh avtorov (Dushanbe, 1972), 59 pp. 

(1965), 43. Ctesias gives no list of satrapies but 94 W. Tomaschek in PW etymologizes the 

rather peoples ruled by the Achaemenids from the name as arta-kauano 'righteous kavi-iike' or 'royal,' 

time of Artaxerxes II, and any projection back- which also does not inspire confidence. 



Achaemenids 113 

723) which tells us nothing. 95 Arachosia (OP Harahuvati) was an important satrapy 
which probably had an Achaemenid garrison town for guarding the southeastern 
frontier of the empire, at least in the later period of the empire when India had fallen 
away from Achaemenid rule. Stephan of Byzantium and other Classical sources knew 
that the land was named after the river, but the capital before Alexander is not 
known, although the fortress of Arshada is mentioned in the Behistun inscription. 
Northeast of Arachosia the hilly region of Sattagydia (OP Oatagu-) and the lowlands 
of Gandhara (OP Gandara) may have been joined to Paropamisos, the center and 
eastern part of modern Afghanistan extending into Pakistan, which was a satrapy 
under the early Achaemenids. 96 It later broke or drifted away from Achaemenid rule, 
and when Alexander came there was no Achaemenid satrap to oppose him. When the 
satrapy of India dropped from Achaemenid rule is impossible to determine but 
probably towards the end of their rule. Gedrosia (OP Maka), or present Baluchistan, 
extending to the west, possibly extending over the water to Oman, was never firmly 
under Achaemenid rule, and it is doubtful whether it ever was a separate satrapy, 
although for Herodotus (III, 94) it is his seventeenth satrapy. 97 

In the west of the empire the satrapies are much better known. The 'Fertile 
Crescent' was at first united under one governor and comprised Babylonia and 
Assyria (which once had extended from northern Iraq to Egypt), and the Arabs of the 
north Syrian desert were included (OP Arbaya). This lasted throughout the reigns of 
Cyrus and Cambyses, and the neo-Babylonian kingdom was retained as a unit with 
two parts officially - Babylonia, and the west or Ebirnari (or Aramaic: Abr Nahara) 
'over the river.' 98 The name 'Assyria' and its derivative 'Syria' have caused much 
difficulty because of changing geographical designations throughout history, but 
when Darius or Xerxes divided the 'Fertile Crescent' into two satrapies the old usage 
of 'Assyria' for the western part, as well as the homeland of Assyria, was retained, as 
we see from the OP lists of Darius and Xerxes; this western part the Greeks and others 
called Syria. Boundaries were adjusted from time to time but the division between 
Mesopotamia and Syria remained to the end of the empire, and even on the relief of 

95 Two etymologies have been proposed for 97 Many attempts to interpret the lists of lands 
Zranka, one 'sea-land' by W. Tomascheck sub and peoples in the OP inscriptions are misleading, 
Drangiane in PW, followed by Herzfeld, and such as G. S. Akhvlediani, "Drevne persidskoe 
another by G. Morgenstierne, comparing Baluchi Maciya-i Gruzinskoe mesx-," Trudy 25 Mezhdun- 
drang 'precipice,' referring to the island mt. of arodnogo Kongressa Vostokovedov, 2 (Moskow, 
Kohe Khwaja in the Hamun lake, NTS, 5 (1932), 1963), 373, where the author identifies Maka as 
43. 'Georgians' in the Caucasus! 

96 Herzteld, op. fir.. 342, identified Sattagydia 98 Cf. O. Leuze, Die Satrapieneinteilung in 
with the Punjab, 'land of seven rivers.' Since the Syrien und im Zweistromlande von 520-320 (Halle, 
figures of the Gandharan, the Sattagydian and the 1935), 1 17-43; H. Bengtson, Kleine Schriften zur 
Indian are almost the same on the reliefs at Allen Geschichte (MUnchen, 1974), 83-100. For a 
Persepolis, showing persons from hot lands (see comparison with the names of satraps at the end of 
Walser, op. cit. [n. 59], 53-55), one would infer the empire, see H. Berve, Das Alexanderreich 
that all inhabited the plains of the subcontinent, (MUnchen, 1926),/fctt«'m. Marquart.op. «'(. [n. 91], 
and probably parts of both the Gandharans and 75, asserted that 'Abr Nahara was called Arbaya 
Sattagydians did, but one must be careful in by the Achaemenids, but the OP lists were of 
drawing conclusions from the reliefs alone. To peoples and not satrapies. 

identify Sattagydia solely with the Punjab is 
unwarranted. The hill country to the west is more 
likely, since it is hot in summer. 



114 Chapter V 

thronebearers on the tomb of Artaxerxes II or III the OP text only speaks of the 
Babylonian and Assyrian, the latter surely for what moderns would call Syrian. One 
capital of the satrapy 'across the river' was probably near Aleppo as we may infer from 
Xenophon, Anabasis (1, 4, 10), or at Tripolis on the sea, although there seem to have 
been several seats of authority for the huge region, Damascus being one of them at 
least for a time." Egypt's boundaries were always fixed by geography, although the 
southern extent of Achaemenid rule up the Nile varied throughout the period of the 
empire. Memphis was the capital, and the Achaemenids were integrated into 
Egyptian culture as pharaohs of a new dynasty. Fortunately we have more documents 
from Egypt (and Babylonia) than elsewhere in the empire, but we cannot transpose 
the Egyptian organization to other parts of the empire. Under the satrap was the head 
of a nome with the Iranian title frataraka, the equivalent of Greek v-napxos, although, 
as mentioned, the latter word was also used for the satrap. Under thefrataraka was the 
chief of a district, the hpthpt' (*haftaxupata) in Aramaic, which has been etymologized 
as the head or protector of one of seven districts, and both of these titles were found in 
Persis after the fall of the Achaemenids. 100 In Egypt perhaps the hpthpt ' was head of a 
topos or toparchy 'county,' although sometimes it seems that thefrataraka was more 
like a toparch. Just as Babylon had revolted in the early years of Xerxes' reign and had 
been brutally suppressed, also Egypt revolted already before Darius died, but Xerxes 
put down the Egyptians and changed the tolerant policy of his father into one of 
control with severity, as Herodotus (VII, 7) says, with Achaemenes, brother of 
Xerxes, as satrap. 

On the reliefs of tribute bearers at Persepolis and in the OP inscriptions appear the 
Ethiopians (OP Kasha) and the Libyans (OP Putaya), but it is uncertain whether they 
were part of the satrapy of Egypt, or more likely just tributaries as Herodotus (III, 13) 
says. In any case, both were on the fringes of the empire and at times did not even send 
tribute to the great king. The people of the north Syrian desert have been mentioned, 
but there is no evidence that the Arabs of Arabia were under Achaemenid rule. The 
Hebrews in Palestine, on the contrary, were usually loyal supporters of the 
Achaemenids even in times of trouble when Egypt or Syria was in revolt, and the Old 
Testament books of Esther, Ezra and Nehemiah, though not to be treated as history, 
indicate a closeness of the two. The diaspora of Jews throughout the empire, from a 
military garrison in Elephantine, an island on the upper Nile, to exiles in Hyrcania, 
probably aided their standing with the Persian overlords. 101 Samaria was a hyparchy 
under the satrapy, and the names of the sub-governors of Samaria from c. 460 to the 



99 Cf. K. Galling, Studien zur Geschichte Israels Egypt under Cambyses was Aryandes, under 
im persischen Zeitaller (Tubingen, 1964), 192, Darius it was Pherendates, for Xerxes his brother 
argues against Aleppo and Damascus as capitals, Achaemenes, then followed the Egyptian revolt 
and for sites on the coast. In Phoencia, the cities of put down by Megabyzes. Under Artaxerxes I we 
Tyre, Sidon, Byblos and Arados were four separate find Arsames as satrap after which a new revolt 
principalities in a special relationship to the empire. broke out in 404 B.C. which led to independence 

100 W. B. Henning, "Ein persischer Titel im until 343 B.C. when Artaxerxes III retook Egypt 
Altaramaischen," In Memoriam Paul Kahle (Berlin, and installed another Pherendates as satrap fol- 
1968), 138—45, and M. N. Bogolyubov, "Titre lowed by Sabakes and Mazakes. 
honorifique d'un chef militaire Achemenide en l0 ' Galling, op. cit. |n. 99], 165-84 on Ezra and 
haute-Egypte," AI, 2 (1 974), 1 09-1 4. The satrap of the Achaemenid governors. 



Achaemenids 115 

end of the empire can be reconstructed thanks to new Aramaic documents. 102 The 
satraps of the land 'across the river' (Syria and Palestine) cannot be determined as in 
the case of Egypt, but some names do appear in the sources such as Tattenai in Ezra (V, 
3) under Darius and part of the reign of Xerxes, Megabyzes under Artaxerxes I, 
Belesys I under Darius II, Abrocomas under Artaxerxes II, ace. to the Anabasis of 
Xenophon 1, 3, 20 (see also Diodorus XIV, 20), although he may have been only the 
sub-governor of Phoenicia, Belesys II under Artaxerxes III (Diodorus XVI, 42) and 
Mazaios, first satrap of Cilicia then of all Syria, under Darius HI. 103 For Asia Minor 
the list of satraps of Lydia and 'on the sea' (Daskylion) is difficult to reconstruct, for 
here, more than anywhere, the joining and separating of territories, plus the 
uncertainty in the use of the term 'satrap' for large provinces and for rulers of 
subdivisions of provinces as well, greatly complicates the picture. After the revolt of 
Cyrus the younger against Artaxerxes II, the province of Cilicia was taken away from 
the ruler Syennesis, and given to a satrap, although provincial coins struck later 
indicate a certain continuation of independent tendencies under the Iranian satraps 
Datames and Mazaios. 104 Noeldeke gave a tentative reconstruction of the hereditary 
post of satrap of Daskylion beginning with Pharnakes followed by Artabazos then 
Pharnabazos, but Mitrobates and Otanes should be added to the list, and the list 
cannot be checked for authenticity. 105 It was different in the important satrapy of 
Lydia, with its capital at Sardis, for Sardis was really the Achaemenid center for Asia 
Minor, and we find a succession of satraps, frequently the same persons as the 
commander-in-chiefs of Achaemenid armies in the west (Greek Kapavos). Beginning 
with Harpagos under Cyrus, Oroites (Herod. Ill, 126) under Cambyses, we find an 
Artaphernes under Darius, perhaps for a time Mardonius under Xerxes, and under 
Artaxerxes I, Pissuthnes, followed by Tissaphernes, Cyrus the younger, then a certain 
Droaphernes who left an inscription probably dating from 367-366 B.C. 106 Then we 
find Tithraustes, Tiribazos, Struthas, Tiribazos again, Autophradates, and under 
Darius III Spithridates, with dates and sequences, as usual, uncertain. In Asia Minor 
there were many local dynasts in Cappadocia, Paphlagonia, Lycia and elsewhere, 
either under satraps or in direct relations with the central government. The 
boundaries of such areas, as well as those of the satrapies themselves, varied 
throughout the history of the empire. From the names of the satraps mentioned 
above, it is clear that Persians, and especially nobles related to the king, were the great 
majority. Indeed, by the end of the empire, the domination of Persians had become 



IO2 /l>M.,210:Mithradates,Rechum,SanballatI, legend tdnmw?); cf. G. F. Hill, Catalogue of the 

his son Delaia, Sanballat II, Hananiah, and Sanbal- Greek Coins in the British Museum, Arabia, 

lat III, to use Galling's transcriptions. Mesopotamia and Persia (London, 1922), cxli; J. 

There are uncertainties of time and extent of de Morgan's reading, op. cit. [n. 103], 50, is wrong. 

rule; cf. Leuze, op. cit. [n. 98], 1 53-63. The identity 105 Article in PW 'Daskyleion' by Rugge. The 

of the coins of %rkmw, supposedly struck at Sinope, names of the satraps of Daskylion and Sardis, as 

and the Abrocomas of Syria is disputed; see J. de well as the commander of Achaemenid forces in 

Morgan, Manuel de Numismatique orientate, 1 the west, are often confused in the sources. 

M " S ' 1923 ~ 26 )- 53 - ° n m e name Tattenai see ' 06 Cf. L. Robert in CRAI (Paris, April 1975), 

io1 Uart ' 0p ' " L f n - 91 1- 52 - 308 - The name *druva-farnah would mean collo- 

For the coins of Mazaios (Aramaic legends quially 'solid luck,' but it might be an appellative 

'* Y) and his predecessor Datames (Aramaic instead of a personal name. 



116 Chapter V 

well established everywhere. One may guess that noble families proliferated and 
intermarried, with family members holding high positions throughout the empire, 
and genealogies of some of the noble families reveal the interconnections. 107 This 
nobility held the real power in the empire, but the bureaucracy was also important, 
especially in regard to financial matters. 

The later Seleucid administrative division of satrapy, hyparchy and toparchy 
probably was inherited from the Achaemenids, and the main function of the 
subdivisions of the province was to collect taxes. 108 Taxes were mostly in kind, even 
after the introduction of coinage in the empire by Darius about 490 B.C. as we learn 
from the Elamite texts of Persepolis. 109 The situation at Persepolis may have been 
unique, since there we find a hierarchy of treasurers, sub-treasurers, those who 
apportioned salaries, others who authorized travel rations, foremen (Elamite 
kurdabattis, OP *gfdapati-) and many others. The impact of the introduction of a 
money economy into the empire must have had far-reaching consequences for the 
growth of trade as well as the stability of the empire, and the reduction of all goods to 
a common money denominator by royal edict was an event in world history of great 
significance. 1 10 Only the great king struck gold coins, called stater by the Greeks, for 
the Iranian word dark (c. 8.4 gm. in weight) was probably named after Darius, rather 
than from a word for 'gold.' Silver was also struck by satraps or by generals for use in 
war, to recruit mercenaries or for rewards, or bribery. If a satrap struck gold coins, as 
in the satraps' rebellion in the time of Artaxerxes II, it was a sign of revolt. Herodotus 
(III, 90 foil.) tells us that taxes were reckoned in talents, a Greek unit of weight, but the 
Iranian system is difficult to reconstruct exactly. A *danaka was 1/8 of a silver shekel 
or aiyXos (c. 5.6 gm.) a word the Greeks borrowed from a Semitic language as did 
the Persians; twenty of them equaled a gold dark, and ten shekels in weight made one 
weight of a karsha, an OP word. 1 1 1 The relation of gold to silver was about 13j :1, 
but it varied as did indeed the weights and measures in various parts of the empire. In 
spite of the introduction of coinage by Darius, it did not develop and receive wide 
acceptance any more than the Old Persian cuneiform writing, for the coins usually 
were accepted only for their metal content and not for any nominal value given to 
them. Later in the empire Greek coinage became popular, even more than the local 
satrapal coins of Cilica, Phoenicia or Asia Minor, such that the Attic coinage became 



107 Cf. F. W. Konig, "Altpersische Adelsge- Dandamaev, "Forced Labour in Achaemenid Iran, 

schlechter," WZKM, 31 (1934), 309; 33 (1936), AF, 5 (1975), 74 

55; 35 (1938), 35; also see F. Justi, Iranisches "° Cameron, supra. Treasury Tablets, 1—4. The 

Namenbuch (Marburg, 1895), 398-99. word *ganza, usually translated as 'treasury,' 

1 ° 8 Usually the OP *hamarakara collected the obviously means 'storeroom' where wine, oil and 

taxes and delivered them to an OP *ganzabara other commodities were stored and not just 

'accountant,' and the *dipi-bara 'scribe' recorded the precious metals or objects. 

transactions. Other words denoting similar func- "' On the relationship of *danaka to a shekel see 

tions occur in various sources, nsgnzsr 'treasurer' in Cameron, op. (it., 37, 1 32. On the value of a karsha 

Aramaic documents. See the various books on see Brandenstein und Mayrhofer, Handbuch, 129, 

Elamite and Aramaic sources noted in the bibh- but especially Driver, supra, Aramaic, 97. On all 

ography at the beginning of the chapter. On the these wts. and denominations see W. Eilers, 

whole the satrapal bureaucratic administration "Akkad. kaspum, Silber, Geld, und Sinnver- 

copied the central one wandtes," Die Welt des Orients, 2 (1957), 326-67, 



109 



See Hinz in ZA, 61 (1971), 274; M. also E. Schwyzer in IF, 49 (1931), 8-28. 



Achaemenids ill 

almost an international tender, while the eastern part of the empire had little to do 
with coins until Alexander. 112 

Taxes were many and varied, on markets, livestock roads, etc., and, of course, on 
land. In Greek sources we find that the revenues of the great king consisted of (popoi 
'tribute' and 8u>pa 'gifts,' or perhaps better 'taxes' and 'tribute,' which probably would 
correspond to OP *bdj- and *hdraka, whereas a 'gift' would be *da$na. 113 Whatever 
the terms used, state income could be classified as from regular taxes plus tribute, and 
from expected gifts, most of which were in kind rather than specie. As far as the 
central government was concerned, state property and the king's private property 
were intertwined, although details of accountancy are lacking. After the conquest of 
Babylonia, and presumably Egypt, Asia Minor and elsewhere, state lands in the 
conquered territories passed into possession of the Achaemenid king, who then gave 
this land to relatives and friends. Under Darius this process was accelerated and not 
only the royal family but the nobility held hereditary estates all over the empire. 
Much of the land was administered by agents such as the banking firm of Murashu in 
Babylonia, although other lands were given in fief (OP bdga) as a reward or payment 
for military service. 114 In Babylonia traditional feudal lands were called after the 
form of service 'bow' land, 'wagon' land, or 'horse' land, and the relation of military 
service to the land tax was complicated, but in any case, land came to be continually 
divided among the heirs of an owner until by the end of the empire military colonists 
no longer supplied any military service and the Achaemenids had to rely on 
mercenaries to fight for them. 1 • 5 In land holding as well as high offices, the Persians 
acquired more and more as time progressed, which cannot have endeared them to 
their subjects, and revolts were more frequent towards the end of the empire. Land, of 
course, was measured, registered and taxed extensively. 

The lot of the ordinary person was very hard in antiquity, also in the Achaemenid 
empire, and in Akkadian documents we hear of parents selling their children into 
slavery to pay debts. 116 Much has been written about slavery in the empire and 

" 2 See the interesting discussion in R. Curiel et Dandamaev, "The Domain-Lands of Achaemenes 

D. Schlumberger, Tresors monltaires d' Afghanistan, in Babylonia," AF, 1 (1974), 123-27, also in 2 

MDAFA, 1 4 (Paris, 1 953), 1 6-30. M. Dandamaev, (1 975), 71-78. On baga- see Driver, op. cit. [n. 1 1 1 ], 

Hisloria, 18 (1972), 45, distinguishes between (1) 40, and Hinz, Alliranisches Sprachgut [n. 79], 53. 

royal gold coins, (2) satrapal silver coins, (3) ' ' 5 For the Akkadian names see Cardascia, op. 

provincial silver coins with the same features as (1 ), cit. [n. 1 1 4], 7, and Dandamaev, "Lehnsbeziehun- 

and (4) local coinage of people conquered by the gen" [n. 48], 41 . The land holders in OP probably 

Achaemenids. were called *dahu-pati-, see Hinz, Altiranisches 

1 n Cf. W. B. Henning, AI. 14 (1977), "Ara- Sprachgul [n. 79], 80, the later dehkans of Sasanian 
bisch harag," 355-57. On *haj- (attested in times. O. Klima, "Zur Problematik der Sklaverei 
Elamite), perhaps originally meaning a 'pot,' see im alten Iran," AF, 5 (1977), 91, considers the 
Hinz, [ch. 4, n. 66], 1 01 , 1 29. On *dahia see Driver, *dahu-pati- a title of one over a subdivision of a 
op. cit. [supra, n. 1 1 1 ], 43, where another word for province, which is unconvincing, 
tax' OP *bara is also discussed on p. 97. Gifts were " 6 Dandamaev, "Kontrakt o sdache vnaem 
usually additions to taxes for special occasions and skota, prinadlezhavshego Satrapu Arshamu," Prob- 
and taxes were assessed on the productivity of the lemy sotsiahto-ekonomicheskoi Istorii drevnego mira, 
land, availability of water and other factors. Sbormk pamyati A. I. Tyumenev (Moscow, 

114 The literature on banking in Mesopotamia 1963), 146, also in detail Dandamaev, "Politische 

■s large; see G. Cardascia, Les archives des Muraia und wirtschaftliche Geschichte," Historia, Ein- 

(Paris, 1951); M. W. Stolper, "The Genealogy of zelschnften, 18 (Wiesbaden, 1972), 33-45. 
the Muralu Family ,"JCS, 28 (1976), 189-200; M 



118 Chapter V 

especially about the meaning of the frequent word in the Elamite documents from 
Persepolis kuttal (OP *gfda-), although originally meaning 'house servant,' in 
Achaemenid times it meant 'worker,' usually subject to corvee labor, but it seems the 
word was also applied to slaves, free men, war prisoners and others, hence a general 
designation. 117 Slave labor was used, especially on the latifundia in Babylonia and 
especially in Egypt, and after the suppression of revolts there was always an increase in 
the number of slaves who were deported to various parts of the empire, a 
continuation of ancient Assyrian practices. Slavery on the Iranian plateau, however, 
does not seem to have been as profitable as in the lowlands. Cuneiform documents, in 
Akkadian or Elamite, tell us about the different wages paid to various workers or 
craftsmen which sometimes varied according to season, demand or geographical 
location. Groups of workers on large projects, such as building Persepolis, were 
organized into tens, hundreds or more, under foremen, with scribes and accountants 
to keep records, evidence of an organized economy. 

Trade, of course, flourished under the empire and luxury objects were carried over 
long distances, with import-export firms such as the house of Egibi in Babylonia 
financing international trade. 1 18 Temples, from ancient times, had been large holders 
of land with slaves, and it seems the temples in Babylonia received a tenth of the 
produce of all who paid taxes before the time of the Achaemenids. 119 Under the 
Achaemenids, however, both in Babylonia, Egypt, and undoubtedly elsewhere, 
although ample sources are lacking, the temples lost their privileged positions, and 
they had to pay taxes and send their slaves on corvee labor or royal projects. Although 
the Achaemenids were tolerant in their regard for all religions in the empire, this did 
not prevent them from levying taxes and other obligations on temples, although 
there were exceptions and certain temples received privileged treatments. 120 

Even though vast sums of gold and silver poured into the royal treasuries, the 
advantages of living in one large empire for merchants and craftsmen caused a 
flourishing of trade and industry. Craft guilds also developed in the towns, much 
more than in the pre-Achaemenid period, and the guilds became centers of security 
and power for their members. Possibly the guilds followed the same organization as 
the associations of feudal tenants called hatru in Akkadian, but not enough evidence is 
at hand to determine guild relationships even though information does exist on their 
workshops, apprentices, patron deities and the like. 1 21 The huge empire did last more 
than two centuries so many people must have supported the government, and we 
may guess that military force was not the main reason for the stability, but perhaps 
one factor in holding the allegiance of the inhabitants of the empire was the legal 
system. 

It is generally recognized that the Achaemenids preserved the past legal heritage in 

117 For various meanings and references see lonien wa'hrend des 6.-4. Jh. v.u Z.," Festschrift fur 
Dandamaev, op. cit. [n. 1 1 6], 39-43, and his Persien F. Altheim, ed. by R. Stiehl, 1 (Berlin, 1 969), 82- 
[supra, n. 20], 189-94, for references. 89. 

118 Dandamaev, Historia [n 116], 50-51. On 12 ° For references see Dandamaev, Historia [n. 
the house of Egibi, whose activities are not attested 1 1 6], 54. 

after Darius I, see S. Weingort, Das Haus Egibi in ' 2 ' D. B. Weisberg, Guild Structure and Political 

neubabylonischen Rechtsurkunden (Berlin, 1939). Allegiance in Early Achaemenid Babylonia (New 

"' Dandamaev, "Der Tempelzehnte in Baby- Haven, 1967), 96-104. 



Achaemenids 119 

the countries they conquered, especially in Babylonia, while the law of that land, just 
as the Aramaic language, spread elsewhere in the empire. The new 'laws of the king' 
(OP data) introduced by Darius made a great impression on his contemporaries, in the 
Bible (Daniel VI, 8, etc., Esther 1, 19), in Plato {Epistles VII, 332 B , Laws, III, Alcibiades, 
etc.), and of course the numerous borrowings of legal terms in Armenian and Syriac 
further attest the great influence of the Achaemenid laws in the Near East. 122 Royal 
judges served in courts to pass on the king's law, according to Herodotus (III, 31 ; V, 
25). The severity of the king's law engendered many stories about punishments, but 
one must be careful in accepting tales as history. There was a dual system of laws in the 
empire, the 'king's law' applicable everywhere, and local laws which were codified by 
order of the king. Akkadian texts speak of the dlnu or the data of the king 
synonymously, one a Semitic, the other an Iranian word, but originally the former 
would be used primarily for local law, and the latter for imperial law, almost a 
distinction between local, religious and secular, and imperial law. 123 Unfortunately 
nothing of an imperial law code has survived and only inferences can be drawn from 
cases which appear in Akkadian or Aramaic sources. Nonetheless the laws of the 
Achaemenid Empire made a profound influence on the ancient world, and the 
Romans, famous for their laws, were building on the precedent of the Achaemenids 
and Alexander. 124 

To turn to the power behind enforcement, the army, one can discern a change 
from the collection of warriors of the Persian people (OP kara) at the beginning of the 
empire to a professional force (spada in Avestan), the core of which were the 
merceneries, at the end of the empire. As with the satraps, Persians came to occupy 
commanding positions in the army, and the officers over Jewish garrison troops in 
Elephantine, Upper Egypt and elsewhere were Iranians. Local troops, of course, were 
mobilized in time of war in various centers (Xenophon, Anabasis, 1,9,7; Oecon. 4, 6). 
The army comprised cavalry (asabara 'horseman') and infantry (pasti- 'footsoldier'), 
and a special part of the foot-soldiers were the archers (OP ©anuvaniya). 1 2S The army 
was divided into units often thousand (OP *baivara-), a thousand, a hundred, ten and 
five (*pasta-dadapati- in Elamite transcription). 1 26 In the Greek sources we hear from 
time to time of a supreme military commander above the satraps, who frequently 

1 22 Plato (Epist. VII, 332B) considers Darius the many others. Cf. Hinz, Altiranisches Sprachgut [n. 
lawgiver whose laws preserved the empire. 79], 86, 97, also R. Schmitt, in Die Sprache, 13 

123 For a bibliography on law see J. Wolski, (1967), 208. For Armenian borrowings, see E. 
Introduction bibliographique a I'histoire du droit et a Benveniste, "Mots d'emprunt iraniens en armen- 
1'ethnologie juridiaue, ed. by J. Gilissen, A/5 ien," BSL, 53 (1957-58), 61-62. 

(Brussels, 1965), 15-17, 23-26. For Akkadian 125 The spearmen were adtika or arStibara, and 

usage see Dandamaev, "Bagasaru ganzabara," in other words appear for short sword and perhaps 

Studien zur Sprachwissenschaft und Kulturkunde, 'battle axe carriers' (vassabara). Cf. G. Widengren, 

Gedenkschrift fUr W. Brandenstein, ed. by M. "Ober einige Probleme in der altpersischen Ge- 

Mayrhofer (Innsbruck, 1968), 237. For royal cases schichte," Festschrift fur Leo Brandt (Cologne, 

in the sources see M. Ehtecham, L'lran sous les 1968), 524-27. 

Achhninides (Fribourg, 1946), 82-84. 126 On the words see E. Benveniste, "Interfir- 

The Iranian words which entered other ences lexicales entre le gothique et l'iranien," BSL, 

languages relating to law are many, especially in 58 (1963), 41-57. See also Aeschylus, 303-05. The 

Armenian. In addition to data we have *databara OP word madiSta 'greatest' used as 'commander' 

judge, frapadaka 'tribunal,' *patifrasa m&frasaka does not imply a military title, 
state's attorney' or 'accuser,' framatar 'judge,' and 



120 Chapter V 

commanded their satrapal troops, and it is uncertain whether this was a standing 
post, or only appointed in time of war. Herodotus (VII, 82—83) names six army 
commanders plus the head of the imperial bodyguard 'the immortals,' as the top 
officers who led the forces assembled by Xerxes into Greece. Whether this meant that 
the empire was divided into that number of permanent military districts or not, we 
cannot determine although it is not impossible. On the other hand, in the army of 
Artaxerxes II against Cyrus the Younger, only four commanders are mentioned 
(Xenophon, Anabasis, 1, 7, 12), which may reflect only those areas subservient to the 
king and not to Cyrus. The first thousand of the 'immortals' were noble Persians who 
served as the imperial bodyguard, who according to Herodotus (VII, 41, also 83) 
carried golden pomegranates on the points of their spears. Their leader was a 
Xihiapxos whose position rose very much after Darius and they are portrayed in 
stone at Persepolis. The archers and cavalry were important for Achaemenid tactics, 
and in open fields they were very effective, but in hilly areas and narrow valleys, the 
Greeks, who wore better armor, usually had the upper hand. 

The garrisons in unruly provinces such as Egypt or Ionia were composed of soldiers 
from all over the empire, to judge from the information from the Aramaic papyri 
from Elephantine, and these troops were settled on land given them as military 
fiefs. 1 27 This practice of settling troops in places other than their homelands, gave to 
the Achaemenids support against local revolts and was a factor in the long life of the 
empire. As time went on, however, the military fiefs were divided and sold and the 
system broke down. As noted, sometimes commanders of armies outranked satraps 
with their provincial troops, but towards the end of the empire the separation of civil 
and military authority was rare and the satraps were more powerful than in the time 
of Darius or Xerxes and even hired their own mercenaries. The fleets were managed 
by the Phoenicians (and Cypriots), Egyptians and coastal peoples of Asia Minor, such 
as the Cilicians and Carians. Soldiers, however, fought on the ships as marines, but in 
naval tactics the Achaemenids were not experienced. In time it became more 
profitable to hire professional mercenaries, mostly Greeks, than to rely on local levies 
whose enthusiasm to serve the Persians declined with economic problems on the land 
which impoverished many, not the least cause of which was the greediness of Persian 
officials and landowners. The armies of the Achaemenids at the end of the empire had 
declined in both spirit and training compared to earlier forces, and devices such as the 
scythed chariots at the battle of Gaugamela did nothing to restore confidence in 
victory, and surely the leadership had faltered in its duty. 

RELIGION 
The question whether the Achaemenids were Zoroastrians or not has brought forth 
much discussion and controversy, but this presupposes the existence of a modern 
concept of a 'Zoroastrian religion' in those days, which is erroneous. 128 Both the 

127 See the interesting remarks in Kraeling, op. Einzelschriften 18 (Wiesbaden, 1972), 59-82. Cf. 
cit. [supra, Brooklyn], 32^18. M. Boyce, Zoroastrians (London, 1979), 48-77, a 

128 J. Duchesne-Guillemin, "Die Religion der summary of her forthcoming, History of Zoroas- 
Achameniden," AA, 19 (1971), 35, surveys the trianism, 2, and the article by G. Gnoli, "Politique 
literature pro and con for this question, and again religieuse et conception de la royaute sous les 
in his "La Religion des Achemenides," in Historia, Achemenides," AI, 2 (1974), 118-90. 



Achaemenids 121 

followers of Zoroaster and the Achaemenids concentrated their worship on the great 
god Ahura Mazda and both did not deny the existence of other deities. Both abhorred 
'the lie' and extolled 'the truth,' as we find in the Gathas and in the OP inscriptions. 
This should be sufficient to indicate that both followed the same religious system, 
although surely with some differences in beliefs if not so much in cult or practices. 
Three general factors can be singled out as the background for discussion about the 
religion of the Achaemenids, first the general Iranian beliefs and practices inherited 
from Indo-Iranian ancestors, second the message of Zoroaster grafted onto, or mixed 
with, the former, and finally ancient Near Eastern religions with temples, priests and 
ancient practices. In time, under the empire the third factor obviously grew in 
importance, the most striking example of which is the remark by Berossos (Frg. Hist. 
Ill, C, 391) that Artaxerxes (presumably II because of his inscriptions) was the first 
Persian king who erected statues of Aphrodite (or Anahita in Iran) in Babylon, Susa 
and Ecbatana. This development under the Achaemenids parallels what we know of 
the religious tradition in the Avesta, and further what one would expect from the 
policy of tolerance towards religions in the empire by the rulers, although religious 
policy was always subordinate to economic and political policy. The best example of 
the dominance of political and economic considerations over others was the harsh 
behavior of Xerxes in Babylonia and Egypt in contrast to his predecessors. The 
destruction of temples by Gaumata and their restoration by Darius possibly refers to 
certain cult centers of the Elamites and other non-Iranian peoples on the Iranian 
plateau, although the details of these actions are unclear. 129 Whether the Magi were 
originally followers of Zoroaster or not is unknown, but they came to be the 
priesthood of the Iranians and at any rate later the preservers of the Zoroastrian 
tradition. 

Only Xerxes among the Achaemenid kings showed strong feelings about religion, 
for not only his destruction of the temple and statue of Marduk in Babylon and his 
actions in Egypt betray this, but his znti-daevic inscription as well. 130 In this he says 
(XPh, 35-41) "among these lands was (one?) where previously daivas were 
worshipped. Afterwards by the will of Ahura Mazda I destroyed that daiva place, and I 
proclaimed 'let the daivas not be worshipped.' Where previously the daivas had been 
worshipped, there I worshipped Ahura Mazda in proper ritual." 1 3 * The identity of the 
daivas has been much disputed, from the gods of Babylonia, presumably proscribed 
after suppression of the revolt, to the Indo-Iranian deities not accepted by 

A. T. Olmstead, History of the Persian Empire Elamites became disillusioned and turned against 

(Chicago, 1948), 93, followed in more detail by the ruler, whereas the Persians at first supported 

Dandamaev, Persien [supra, n. 201. 234-37, claim Bardiya and then the pseudo-Bardiya VahyazdSta. 
that Gaumata destroyed the temples of the Persian 130 For a description of his actions and the 

nobility, even though there is no evidence that the sources, see Olmstead, op. cit. [n. 1 29], 235-37, and 

Persian nobility had temples or cult buildings Dandamav, Persien [supra, n. 20], 240, n. 1088. 
whereas the Persian people did not. An indication l31 The last two words artaca-brazmaniy have 

that both in Persis and in Elam the Elamites first had many interpretations; cf. J. Duchesne-Guille- 

supported Darius and then turned against him in min in BSOAS, 25 (1962), 336-37, who suggests 

the Behistun inscription are the revolts of Assina 'facing ana' or 'a«a-wards.' Whatever the interpre- 

and Martiya, both of whom were deserted by' the tation, it means 'in proper ritual and attitude.' I had 

Elamites, although later they supported Atamaita. proposed 'properly with the law (arta)' in my 

en though Darius restored their temples, some Heritage [supra, ch. 3, n. 26], 127. 



122 Chapter V 

Zoroastrians. 132 Xerxes was acting in a more aggressively religious manner than his 
predecessors which does not, however, prove that he was in any way more orthodox' 
Zoroastrian than other rulers. No matter what the beliefs of the rulers, from the 
Elamite documents at Persepolis, we see that thoughout the reigns of Darius and 
Xerxes the deities worshipped in the area were Iranian, Elamite and even Babylonian, 
as well as rivers and mountains, offerings to all of which were subsidized by the state. 
We find a priest called Marduniya (Mardonius, a good Iranian name) who sacrifices 
to the Elamite god Humban (PFT 348), while a magus with the un-Iranian sounding 
name of Ukpis received rations for a ceremony dedicated to Mithra, a mountain and a 
river. ' 33 The ceremony called Ian in the texts, according to some, was reserved for the 
god Ahura Mazda but this is refuted by the texts themselves where syncretism and 
multi-faceted worship is found. 1 34 It is not possible to draw general conclusions from 
the presence or absence of designations of persons in the tablets; for example, of the 
nine persons called magus two have names which are difficult to identify as Iranian, 
viz. PFT 1798 Limepirde and 1955 Ukpis, while 757 and 2036 Kurka, also seems 
difficult to claim that he was Iranian. Among those designated as 'priest' (Elamite 
Satin), too we find names obviously Iranian, such as Mardonius (PFT 348), sacrificing 
to Humban, the chief Elamite deity. So what conclusions can one draw from the 
Elamite texts about religions in Fars province? The king, and presumably much of 
the court, worshipped Ahura Mazda at least in word if not in deed. But the rulers also 
not only tolerated, but had rations given for ceremonies and libations to many deities, 
including rivers and mountains, honored by the local inhabitants or foreign workers 
at Persepolis. Two words appear in the texts for priests, one Iranian magus and the 
other Elamite Satin, but their functions seem to have been very similar, and it is clear 
there were many religious libations and ceremonies in the daily life of the people, 
Elamites and Persians, with no great conflicts between them. One may suppose, 



132 For the Babylonian thesis see Duchesne- ' 34 The meaning of the Ian ceremony is 
Guillemin in HiitoriVi, Einzelschriften, Heft 18,66, unknown, but it seems to have been a general 
and for the Slammesgollheiten of the Iranians see religious ceremony with offerings in which all 
Dandamaev, Persien [supra, n. 20], 239, who calls participated, although it may have been Elamite in 
Xerxes a follower of Bardiya in this regard origin. Cf. H. Koch, Die religiosen Verhattnisse der 
Incidentally the old Babylonian triad of gods - Dareioszeit (Wiesbaden, 1977), 181. Unfortunate- 
Anu, Nabo and Istar - mentioned by Gnoli (in ly, Koch is mistaken in various details, e.g. p. 178 
Duchesne-Guillemin, op. cit , 68, and in AA, 19 that Mithra does not appear in the tablets (sic), and 
[ 1 97 1 ], 29) as a parallel with Iran is highly dubious 1 58, n. 200, where she implies that Persian Magi 
as any specialist m Mesopotamian religions knows. only worshipped Ahura Mazdi, but in PFT 1957, 
The best explanation of the daiva worshippers is Irdazana a Persian magus conducts a religious 
either aberrant Indo-Iraman cults to the east of ceremony for the Elamite deity Turma and Mithra 
Persis, or the Elamites in Khuzistan (and Persis), not for the Ian ceremony, while another magus 
but not Babylonia which revolted and would have (PFT 1955) received rations for Mithra and for a 
been described in the inscription in the usual mountain and a river. Furthermore to say (p. 138) 
fashion. that no Ian ceremony was performed for Elamite 

133 References are to Hallock, Persepolis Forlifi- deities is incorrect, e.g., see tablet no. 773. Thus 
cation Tablets, [supra, Lit.], (PFT). See also my offerings are made to many spirits, which is 
article on "Religion in Fars at the time of the characteristic of folk religions. 
Achaememds" (in Persian) in Chaharomln kongreye 

taltqlqat-e Irani, ed. by M. H. IskandarT (Shiraz, 
1353/1975) 2, 218-21. 



Achaemenids 123 

however, that gradually the Iranian deities, with the great Ahura Mazda above all, 
replaced the others and Iranian rites and rituals became dominant. 

Burial customs always have been an index of religion, although interpretations 
based solely on archaeological finds must be supported by other evidence. All dead 
bodies decay, and the disposal of them is incumbent on all societies. Cremation, 
mummification, exposure or burial were the usual methods of disposal, all of which 
removed decaying matter from the living, thus avoiding pollution. Gradually the 
practice of exposure of the dead, which Herodotus (1, 140) says was practiced mainly 
by the Magi, came to be the accepted way of disposal, although others (presumably 
royalty and nobility) were covered with wax before being buried. Some scholars 
suggest that the rite of exposure and the collection of the bones in an astodan or 
ossuary was native to eastern Iran and Central Asia and then was gradually adopted in 
western Iran, although even in Asia Minor of the Achaemenid period archaeological 
evidence of this practice is found. 135 In any case, even the embalming of bodies and 
placing them in stone receptacles was considered adherence to a belief not to pollute 
the earth with decaying matter, and the mode of disposal of the dead should not be 
used to prove or disprove the Zoroastrian belief of the defunct. We know where the 
Achaemenid kings were interred at Naqsh-e Rustam and at Persepolis in rock-cut 
tombs, with rock-cut places for family members covered with stone covers. 

When we turn to the Aramaic inscriptions on the mortars and pestles and plates of 
green chert found at Persepolis we find a picture different from the Elamite tablets, 
for the names seem to be all Iranian, and a certain fondness for the theophoric names 
in "Mithra", at least much more relatively than in the Elamite tablets, indicates at least 
that Mithra was not proscribed at Persepolis. The inscriptions, however, are merely 
records of registration of the objects, possibly cut or fashioned in different rooms of 
the 'treasury' at Persepolis. 136 What is clear, however, is that these objects in green 
stone were made for cultic purposes rather than for culinary aims since the numerous 
mortars and pestles were hardly used just for crushing seeds or nuts. On the other 
hand, they do not seem to have been used much, if at all, and the Aramaic inscriptions 
written in ink were placed mostly on the heads of pestles, an unlikely place for an 
inscription on an object to be used. 137 In other words, the stone which came from 
Arachosia in eastern Iran may have been made into mortars, pestles, plates and trays 
by Arachosians as presents or tribute for the Achaemenid king who had them 
registered and deposited in the 'treasury' of Persepolis, but never used them. So the 
promise of any new material on the religious picture at Persepolis in the inscriptions 
on these objects seems to have evaporated. 

Fire probably was an important feature of religious ceremonies for the Iranians as 
well as for the Elamites who had special priests for the fire ceremonies. 138 Since at 

S. Shahbazi, The Irano-Lycian Monuments Ritual Texts from Persepolis' de R. A. Bowman," 

(Tehran, 1975). 125-27 and 154-56. Xanthus in AI, 2 (1974), 193-217. 

(FHG I, 42; Frg. Hist.) says that Persians claim l37 Bowman, op. cit. [supra, Aramaic], 1-5, 16, 

they learned from Zoroaster the rule against 44-52. 



burning dead bodies or defiling fire, which may be ' 38 On the Elamite 'fire' cult see F. W. Konig, 

cor ^ e 6 ct - Die elamischen KSnigsinschriften, Archiv far 

Cf. the book by Bowman [supra, Aramaic], Orientforschung, 16 (Graz, 1965), 58, and the 

and the convincing interpretation of the inscrip- 'fire watcher' of the Persepolis treasury tablet, 

tions by J. Delaunay, "A propos des 'Aramaic Cameron, PTT [supra], 7, no. Fort, 3126. 



124 Chapter V 

Persepolis, no temple or altars have been found, we may conclude that on the whole 
the court did not pay much attention to religion, and that royal religious ceremonies 
were performed on smoothed platforms in the stone in the open on the heights above 
the platform, as Herodotus (I, 131) said of the Persian sacrifices. The answer to the 
question regarding the religion of the Achaemenids is that it was in the general Mazda" 
worshipping framework of which Zoroaster himself was a part, but towards the end 
of the empire many influences, especially Babylonian, possibly because of the 
Babylonian mothers and concubines of later kings, came into Achaemenid beliefs and 
practices. But certain priests maintained what they thought were proper rites and 
rituals, and there may have been a school of A vesta learning at Persepolis which both 
resisted and compromised with other beliefs and practices. How much influence 
priests who called themselves followers of Zoroaster had upon royal pupils is 
unknown. 139 One may speculate that by the time of Alexander the Great the Magi 
had become the 'Zoroastrian priests' for many if not most Iranians in the empire, and 
had begun the amalgamation process which was to result in the fixed text of the 
A vesta and the rites and rituals as we know them in Sasanian times. 



THE CAPITALS 

The first capital of the Achaemenids was Pasargadae where Cyrus lived in palaces 
surrounded by parks and gardens with a citadel or platformed area about which we 
know much, thanks to excavations. 140 Much has been written about the palaces, the 
tomb of Cyrus and the enigmatic ruined building called the Zendan which has a copy 
at Naqsh-e Rustam near Persepolis. These two structures have been designated as fire 
temples, tombs, or archives where copies of the Avesta or royal paraphernalia, 
perhaps for coronation ceremonies, were kept, but no conclusive evidence about their 
function has been found. 141 The theory that they were both fire temples has been 
almost abandoned, since no fire temple of a similar structure can be found for 
comparison, and the stone doors and lack of windows make any internal fire altar 
most dubious. Since deep traces of much usage of the doors on both are evident, the 
theory of a tomb does not appear likely, so we are left with a structure either where a 
dead body was prepared for last rites or it was a kind of safe for objects. In either case it 
was reserved for royalty, and since the Zendan is said to be older in date than its 
counterpart the Ka'bah of Zoroaster at Naqsh-e Rustam because of the use of tooth 
chisels on the latter, but not on the former, one may suggest that the structure at 
Naqsh-e Rustam was erected by Darius after moving his 'capital' from Pasargadae to 
Persepolis. 142 

139 Plato in Alcibiades I, 122, says a future king C. Nylander, Ionians at Pasargadae (Uppsala, 
was taught Magian lore by Zoroaster son of 1970); and A. Sami, Pasargadae (in English), 
Oromaz. On the 'Avestan' school in Persis see K. (Shiraz, 1971). 

Hoffmann, "Das Awesta in der Persis," Pro/egoitif/id m On the various theories see E. Schmidt, 

to the Sources on the History of Pre-lslamic Central Persepolis, 3 (Chicago, 1970), 34-49, and Stronach, 

Asia, ed. by J. Harmatta (Budapest, 1979), 89-93; op. cit. [n. 1], 132-36. 

also my "Religion in Fars under the Achaemenids", l42 On the chisel marks see C. Nylander, "The 

in Milanges pour J. Duchesne-Guilletnin, ed. by P. Toothed Chisel in Pasargadae," AJA, 70 (1966), 

Lecoq (Lou vain, 1984) 373-76, and Stronach, op. cit. [n. 1], 99-100. 

1 40 See Stronach, Pasargadae, op. cit., n. 1 , passim ; 



Achaemenids 125 

Following archaeological discoveries there is good reason to believe that on the 
plain of Marvdasht, a number of palaces had been built with gardens and parks similar 
to Pasargadae before Darius conceived Persepolis. 143 Having seen the architectural 
splendors of Babylonia and especially Egypt, Darius apparently resolved to build a 
magnificant complex of palaces on a large platform as a summer capital, but also as a 
royal symbol of the power and glory of the Persians in their homeland. Darius started 
building on the platform c. 520 B.C. and work continued on it until the end of the 
dynasty, although the first three rulers did the most. 144 The platform was called a 
fortress (in the Elamite tablets birta, OP dlda), but it was more a ceremonial area than a 
military center, for the king and his courtiers probably lived most of the time in the 
palaces at the foot of the acropolis when they stayed in Persis. The ruins of Persepolis 
have provided material for innumerable books and articles on the art, architecture, 
royal ideology and much more. The whole area was surely occupied and on the 
platform embassies were received and all who came were impressed, for such was the 
intention of the builders, the Achaemenid kings. We cannot discuss the many theories 
that Persepolis was constructed according to certain astronomical orientations, or that 
it was solely a ritual city, for these sometimes fascinating observations have little 
bearing on history. 14S Likewise the implications of OP words relating to buildings at 
Persepolis (and Susa) such as talara 'summer palace,' hadii 'royal residence, or seat' 
apadana 'courtyard,' and others have produced voluminous writings which have 
enriched our knowledge of Achaemenid times but again cannot be discussed here. 146 
The false idea that the Greeks knew nothing about Persepolis before Alexander is 
refuted by Ctesias who uses the name Persis for both the site and the province, and by 
the notices of Strabo 727-28, Diodorus XVII, 68-71, and many others who, though 
later writers, quote earlier sources. 

Susa has been excavated by French archaeologists for almost a century and much 
has been learned about the winter capital of the Achaemenids, which rich city was 
probably so designated first by Darius. 147 The uncovering of new palaces and the first 
Achaemenid sculpture in the round, of Darius, has provided much new material for 
reconstructing our picture of this winter capital. At both Susa and Persepolis many 
artisans and workmen from all over the empire constructed and decorated the 
buildings; perhaps never since the ancient pharaohs of Egypt had so many workers 
labored on the construction and embellishment of royal structures, and an inscription 
gives details about the precious stones, cedar from Lebanon, and other materials used 
in construction and which people worked on the various projects (Darius, Sf). 
Babylon was another place of residence in the winter and Ecbatana in the summer, 



43 A. B. Tilia, Studies and Restorations at and S. Shahbazi's observations on the number of 

Persepolis and Other Sites of Fars, 2 (Rome, 1 978) steps in the great staircase and other measurements. 

73-91 • 146 On these terms see W. Eilers, "Die Ausgra- 

144 Cf. A. S. Shahbazi, "New Aspects of bungen in Persepolis," ZA, N. F., 19 (53) (Berlin, 
Persepolitan Studies," Gymnasium, 85 (Heidelberg, 1959), 248-60; G. Ito, "On the Function of Darius' 
1978), 487-500, and his "From Parsa to Taxt-e Palace (talara) at Persepolis," Memoirs of the Faculty 
Jamsld," AMI, 10 (1977), 197-207. of Letters, Kyoto University, 13 (1971), 1-23. 

145 W.Lentzund W.Schlosser, "Persepolis -ein 147 Arrian III, 16, 7, and Curtius V, 2, 10-12, 
Beitrag zur Funktionsbestimmung," ZDMG, 121 and the many volumes of the Delegation archeolo- 
(1971), 254-68, also talks by A. U.PopeJ. George, gique Francaise en Iran, as well as the Cahiers. 



126 Chapter V 

but neither one had money lavished on them as did Persepolis and Susa, both of which 
were particularly symbolic of Achaemenid power and majesty. 

XERXES AND STAGNATION 

The reign of Xerxes was a period of consolidation of the foundations built by his 
father, but it was also a change of direction in both the religious and ruling policies of 
his predecessors, and the beginning of a stagnation and decline in various features of 
the Achaemenid Empire. In an inscription at Persepolis (XPf) Xerxes, whose throne 
name seems to mean 'ruling over heroes,' says that Darius had other sons, but his father 
made him greatest after the king, and in this and other inscriptions it is clear that 
Xerxes was very much in the shadow of, and under the influence of, his father. 148 
According to Herodotus (VII, 2) and Aeschylus (753), Xerxes was the son of Atossa, 
the daughter of Cyrus, whom Darius married after coming to power. It is idle to 
speculate on harem intrigues which may have induced Darius to select Xerxes as his 
successor, but it is significant that he made Xerxes his crown prince, and presumably 
his representative in Babylon, although cuneiform documents do not reveal this. 
They do tell us that Darius probably died in November, 486, B.C., and Xerxes 
succeeded him, but a few months earlier Egypt had revolted from Achaemenid rule 
and Xerxes had to re-establish control over that rebellious satrapy before he 
embarked against Greece. 149 Xerxes was successful in Egypt and changed the policy 
of his father, who had subsidized and built temples and had employed Egyptians in 
high positions. Xerxes apparently stopped all of these, and Herodotus (VII, 7) says he 
laid a heavier yoke on the Egyptians than his predecessors. 

Babylonia followed Egypt in revolt, the reasons for which are unclear. Cuneiform 
tablets in Akkadian indicate two rebels were recognized as rulers in Babylon in the 
fourth year of Xerxes, one called Bel-shimanni and the other Shamash-eriba. 150 
Xerxes was able to suppress these revolts, and he changed the traditional titles on 
documents there: 'king of Babylon, king of lands,' to an additional 'king of Persis and 
Media,' an indication that the Achaemenid ruler had changed his attitude towards 
Babylonia, and although some modern scholars describe Xerxes' destruction of 
Babylon in vivid detail the sources are silent about this. In any case, Xerxes is most 
remembered for his invasion of Greece, which began in the spring of 480 B.C. The 
battles of Thermopylae, Salamis, and Plataea are well known, and much has been 
written about them based on Classical sources. There is nothing to add from the 
Persian side, except to caution against the inflated numbers of the Achaemenid forces 
given in Greek sources. The military discipline and better armament of the Greeks 
were important factors in their victories. The later invasion of Asia Minor in the late 
summer of 479 and the decisive defeat of Achaemenid forces at Mykale ended the first 

148 We do not know the personal name of Kienitz, op. cit. [n. 36], 67. An analysis of the 
Xerxes. For an assessment of his character see M. reaction of Greeks to Xerxes' invasion of Greece in 
May rhofer, "Xerxes, K6nigderKonige"-4/ma«ac/i their literary writings is given by W. Kierdorf, 
der Osterreichischen Akademie der Wissenschaflen, Erlebnis und Darstellung der Perserkrieg, Hypomne- 
119 (Vienna, 1969), 158-70. mata 16 (Gattingen, 1966), 130 pp. 

149 For dates of rule of the kings see Parker and lso Parker, op. cit. [n. 56], 17. 
Dubberstein, op. cit. [n. 64], 17-19. On Egypt see 



Achaemenids 127 

phase of Greek-Persian hostilities, and the Persians, now on the defensive, turned to 
intrigue and bribery to repair their blunders in the field. 

We hear no more of Xerxes either in the field, as on the expedition against Greece, 
or in new endeavors. The book of Esther in the Bible and Aeschylus' Persians give an 
unflattering portrayal of Xerxes, who is said to have retired to his harem after the 
defeat of the Persians at the hands of the Greeks. The politics of Themistocles, 
Pausanius and others, accused of being pro-Persian, is part of Greek history and need 
not detain us here. The steadily growing Greek cultural influence in the western part 
of the Achaemenid Empire, especially in Anatolia, can be seen in monuments, in 
inscriptions and art objects, while Greek craftsmen were active at Persepolis, where 
Xerxes did so much construction, and elsewhere. 151 Xerxes was probably in 
Persepolis when he was murdered in the summer of 465 B.C. 

Ctesias (29-30) gives an account of the murder of Xerxes by a certain Artabanus 
and a eunuch, Spamitres, who then told one son Artaxerxes (whose personal name 
was Cyrus according to Josephus (Jewish Antiquities, XI, 6, 1) that Xerxes had been 
killed by his brother Darius, the crown prince. Artaxerxes then had his brother killed 
and assumed the throne, apparently after dispatching both Artabanus and 
Spamitres. 152 Probably an uprising broke out at the death of Xerxes, reported by 
various Classical sources as that of a son of Xerxes in Bactria, but which was quickly 
suppressed. 153 The earliest cuneiform tablets dating from the reign of Artaxerxes 
(throne name meaning 'rule by rectitude') begin in June 464 and end in December 
424, a long reign. Since Herodotus does not extend into the reign of Artaxerxes we 
are left with only a few scattered fragments. Artaxerxes, unlike his father, left few 
remains in Persepolis, and apparently he spent more time in Susa and possibly 
elsewhere on the plateau. 

Egypt revolted again in 460 and a certain Inarus, who had been in control of Libya, 
soon became ruler of most of the country of Egypt, and for the account of events we 
rely on Ctesias (32-35), who unfortunately is hardly a reliable source. 154 The 
Athenians aided Inarus, but Megabyzus, a noble Persian, with the aid of some 

151 Cf. Nylander, op. cit. [n. 140]; J. Hofstetter, Treasury in Achaemenid times, I perhaps too 
Die Griechen in Persien (Berlin, 1978). boldly had suggested that Darius had been 

152 S. Shahbazi, "The Persepolis Treasury Re- involved in the murder of his father because of 
Hefs Once More," AMI, 9 (1976), 151, claimed to Herodotus (IX, 108) explained by Olmstead, op. 
have first proposed that the bas-relief of a seated cit. [n. 129], 289. Shahbazi rejects all of this, 
king and his crown prince seated behind him in the presumably as Greek calumny. It is difficult to 
treasury of Persepolis represented Xerxes and his know which source is telling the truth and rather 
crown-prince Darius, rather than Darius and than dogmatically assert my own views as right, 
Xerxes as had been hitherto assumed. In reality, it would be better to reserve judgment since we 
first, both myself and H. von Gall independently do not know. Shahbazi's confident assertion that 
came to this conclusion before Shahbazi, and Artaxerxes had no part in the murder of his 
second, his further reconstructions and identifica- brother is based on the assumption that Iranian 
tions in the article above are subject to doubt. For rulers would not stoop to fratricide. For a later 
example, Diodorus (XI, 69) says Artaxerxes slew dating of the transfer of the relics see P. Calmeyer, 
his brother Darius, although here Shahbazi rejects AMI, 9 (1975), 78-79. 

Diodorus while accepting him elsewhere. In my ' " References in Olmstead, op. cit. [n. 129], 290, 

article, "Persepolis Again," JNES, 4 (1974), 383- n. 3. 

86, submitted to the journal in 1 972, shortly after IS4 J. M. Bigwood, "Ctesias' Account of the 

the discovery by G. Tilia that the two reliefs had Revolt of Inarus," Phoenix, 30 (Toronto, 1976), 19. 
been removed from the Apadana and put in the 



128 Chapter V 

Egyptians who had not supported the revolt according to Thucydides (I, 104, 2), 
reconquered Egypt. The new satrap of Egypt Arsames (O. P. ArsSma) helped to defeat 
the Athenian fleet which had come to the assistance of the Egyptians, and by 454 B.C. 
Egypt was again in Persian hands, but not completely or firmly, and it was only by 
the peace of Kallias between Athens and the Persians that Greek interference in Egypt 
was ended. Nonetheless in Libya and the western delta some Egyptians still 
maintained independent rule not under Persian jurisdiction, as we learn from 
Herodotus (II, 30 and III, 15) who visited Egypt at this time. Egypt was a center of 
unrest throughout Achaemenid rule, and since it was located far from Persis, 
opportunities for independent actions of the Egyptians were many, but information is 
very sparse throughout the fifth century B.C. Nonetheless, throughout the rule of 
Artaxerxes I it seems that Egypt on the whole remained loyal to the Achaemenids and 
not until 404 B.C. in the time of Artaxerxes II did Egypt successfully revolt and 
maintain independence for some sixty years. 

The peace of Kallias, named after the Athenian ambassador to the Persians, has been 
mentioned, and this was an important event in the reign of Artaxerxes I, which 
established a modus vivendi between the Ionian cities under the Delian league, which 
was really an Athenian empire, and those under Persian rule. This peace in 448 B.C. 
gave a certain prosperity to both sides, with an agreed non-interference by each in the 
affairs of the other. 155 Also during the reign of Artaxerxes I may be placed the 
rebuilding of the temple in Jerusalem by Nehemiah, but one must be careful in 
accepting much of the information in the Bible as fact, especially when it refers to the 
royal court and the roles of various people in it even though certain details, as in the 
book of Esther, betray more than a casual acquaintance with affairs of court. There is 
no point here in examining the stories of court intrigue reported by Ctesias, for even 
though interesting they only reveal the decadence and greed of the royal family. 
Records from Babylonia show the increasing control of land by the Persian nobility, a 
situation undoubtedly matched elsewhere in the empire. As long as tribute and gifts 
came to the Achaemenid court, however, the satraps were allowed to rule practically 
with free hands. Anatolia, on the other hand, was especially important for the 
Achaemenids, since the Greeks were enemies and when Tissaphernes was made 
commander of all forces on the Aegean Sea coast, he was above the satraps in rank. In 
408 B.C. he concluded a truce with Athens indicating his position over that of the 
satraps, a situation, however, applicable only to Ionia, or perhaps to other frontiers of 
the empire where a supreme military command was necessary. 

It should be borne in mind that our sources are Greek and they naturally place great 
emphasis on matters of interest to the Greeks, and it is only by chance that a remark is 
made that throws light upon internal affairs of the empire. We know about frequent 
revolts in Anatolia, Cyprus and Egypt, but the corresponding picture of a quiet and 
monolithic Iran is hardly true. A revolt of the Medes occurred at the beginning of the 
reign of Darius II according to Xenophon (Heltenica 1, 2, 19), while the Cadusians on 
the Caspian Sea revolted about 405 B.C. (II, 8, 13), and we may surmise that many 

1 55 The terms of the treaty are discussed in detail Heichelheim, "Geschichte Kleinasiens" in HO, 2, 
by Olmstead, op. cit. [n. 1 29], 311. The date of the Keilschriftforschung und alte Geschichte Vordera- 
treaty, 448 or 449 B.C., is in dispute; cf. M siens (1966), 35-36. 



Achaemenids 129 

other internal revolts took place about which we hear nothing. From Babylonian 
cuneiform tablets we can infer the general and steady economic decline and the 
growing impoverishment of the people. With high interest rates of 40 or 50 percent 
on loans, usury was rampant, and many were the victims of both the ruinous taxation 
and the special levies for the suppression of revolts, or gifts to the royal court. The 
Babylonian and Egyptian (mostly Aramaic) documents tell us of ordinary life, 
whereas the Greek historical sources on the Persians are primarily about court 
intrigues or details of battles, with little concern for the ordinary subjects of the 
empire. Consequently it is very difficult to construct a picture of ancient Iran on the 
basis of two kinds of foreign sources. Furthermore we can never determine whether 
conditions in Babylonia and Egypt also obtained in the heartland of the Achaemenid 
empire, and it would be dangerous to infer that the conditions of land tenure, taxation 
and the like, in Mesopotamia also applied to Persis, the Achaemenid homeland, thus 
we are reduced to conjecture and surmise. We may be certain that taxes such as tolls 
on roads, market taxes on animals, and many others were manifold and oppressive 
everywhere. 

The work of M. Dandamaev on Babylonian cuneiform documents of the 
Achaemenid period indeed has given us a picture of economic and social conditions 
in Mesopotamia, especially in regard to slavery and forced labor under the 
Achaemenids. 156 The result of this research is that various kinds of slavery and forced 
labor existed in different parts of the empire, and the Elamite word kurtai in the 
cuneiform documents meant 'workers' in general but under different conditions, 
many serving a term of forced labor on royal estates, others enslaved prisoners of war, 
and even free workers earning wages. It is interesting that at Persepolis under Xerxes 
the kurtaH received payment according to the work done irrespective of their legal 
status, and the rewards here were higher than elsewhere. 157 Land ownership under 
the later Achaemenids also underwent a change from the earlier periods. The 
granting of 'feudal' lands to officers and soldiers, mentioned in the sources as 'bow 
land,' 'horse land' and 'wagon land,' began soon after the Achaemenid occupation of 
Babylonia, but the constant division of those lands among heirs impoverished the 
descendants of the military fief holders. By the end of the fifth century the military 
obligations of these settlers had fallen in abeyance, and mercenaries had taken over 
much of the military needs of the Achaemenid empire. 1 58 Obviously there are many 
details which cannot be discussed in a general book, but the picture we have of 
impoverishment of land holders in Babylonia probably is true of most of the empire 
in the last century of its existence. 

The involved system of tax collecting in Babylonia, where feudal obligations and 
other taxes were managed by banking firms, such as Murashu and sons at Nippur, the 
Egibi family in Babylon and others, is fascinating in revealing an economic life based 
on credit and loans with modern features. When we remember that the introduction 

1 56 See especially a series of his articles in AF, 1 Sklaverei im alten Iran," AF, 5 (Berlin, 1977), 91- 

(1974), "The Domainlands of Achaemenes in 96. 

Babylonia," 123-27, and "Forced Labour in 157 Dandamaev, AF, 2 (1975), 78. 

Achaemenid Iran," 2 (1975), 71-78, and his book 158 Dandamaev, ..Lehnsbeziehungen" [n. 48], 

Rabstvo v Vavilonii VII-IV vv. do n.e. (Moscow, 37-42. 
1974), Cf. also O. Klima, "Zur Problematik der 



130 Chapter V 

of payments in money, indeed coinage itself, is an Achaemenid phenomenon, then 
the significance of greatly expanded commercial and banking activities can be 
appreciated. It is not too much to suggest that the spread of a money economy played 
an important role in the stability of the Achaemenid empire and continuing 
allegiance to it. The traditional explanation of the fall of the empire as the result of 
abuses of their positions by those in power, the decadence and corruption at court and 
among the aristocracy, combined with a fall in the standards of living of the common 
folk, can be further documented by Babylonian tablets. 159 The ever growing 
taxation and the greed of Achaemenid officials, just to mention two factors, helped to 
undermine the expansion of agriculture and irrigation, trade and commerce and 
handicraft production, which had characterized the early pax Achaemenica, which had 
provided prosperity for many subjects. 

DECLINE OF THE ACHAEMENIDS 

The first half of the fourth century B.C. is almost a complete blank in our history of 
Iran, and of the entire empire. Inscriptions are rare and Classical sources also reflect 
little of the internal affairs of the empire, while the Babylonian documents cease. This 
latter situation represents a change in recording from clay tablets to papyri, leather, 
or ostraca with Aramaic writing instead of cuneiform, which came to be restricted to 
temples and the activity of priests who copied and recopied ancient spells and charms 
or religious texts. Scholarly articles dealing with Babylonia in the late Achaemenid 
period are characteristically scant in the information they provide and without 
sources one is reduced to conjecture. 160 Likewise Egyptian, Aramaic and Anatolian 
(primarily Lydian and Lycian) inscriptions give us little new information which can 
be used for historical purposes. One development of Achaemenid weakness, though 
nowhere expressly stated in sources, was the increasingly independent actions of 
satraps, especially those in Anatolia. One has the impression that they negotiated with 
Greeks much as they pleased, with or without royal sanction, which was sought only 
when needed. The varying fortunes of Athens and Sparta and the activities of the 
satraps of Sardis following Tissaphernes - Tithraustes, Tiribazos, and Autophradates - 
again are part of Greek or Anatolian history and shed little light on inner-Iranian 
history. 

Egypt became independent from the Persians about the time of the revolt of Cyrus 
the younger, and it remained so until 343-342 B.C. when Artaxerxes III reconquered 
it. Even though another 'King's peace,' the peace of Antalkidas (the Spartan envoy to 
Artaxerxes II) in 387 B.C. had left the Ionian cities of Anatolia and Cyprus to the 
Achaemenids, the cultural and economic influence of Greece greatly increased 
throughout Anatolia, and the political control of the Achaemenids was greatly 
weakened compared to the previous century. The shifting of policy from friendship 
with Sparta to Athens under Konon meant neither a strengthening nor a weakening 
of Persian power but rather a full participation in the ever-shifting fortunes of Greek 

159 Dandamaev, Rabslvo [n. 156], esp. 28-44. 1965), 330-55. Even the cuneiform sources of the 

160 For example, M. Meuleau, "Mesopotamien house of Murashu cease after Artaxerxes II, and in 
in der Perserzeit," in H. Bengtson, ed., Criechen und general we have very little from Babylonia in the 
Perser, Fischer Weltgeschichte, 5 (Frankfurt/M, later Achaemenid period. 



Achaemenids 131 

politics, as another 'King's peace' of 371 indicates. The real weakening of Achaemenid 
power, however, came from the independent actions of the satraps in Anatolia. 161 

About 368-367 B.C. the satrap of Cappadocia, Datames, openly proclaimed his 
independence from central authority, but the Classical sources do not give a clear 
picture of the causes and the course of the rebellion. The time-honored practice of the 
court, setting one satrap against another, after initial success in the end failed, and 
Ariobarzanes, satrap in Daskylion, joined Datames in revolt, and still another revolt 
broke out when Aroandas (Orontes), satrap of Armenia, was ordered to go to Mysia. 
Autophradates, satrap at Sardis, kept his loyalty to Artaxerxes II, but the success of the 
rebels led him too to change sides, and he was joined by local dynasts such as Mausolos 
of Caria. Diodorus (XV, 90) claims that the coalition included all of Anatolia, the 
Spartans, Egyptians, Phoenicians, and others. Aroandas became the leader of the 
rebels, but then he betrayed his colleagues to the king and the revolt collapsed. Even 
the Egyptian rebel ruler Takhos was overthrown by the Egyptians and made his way 
to the Achaemenid court where he was forgiven (XV, 92). In the fighting between 
satraps Greek mercenaries were the main forces, an indication of the decline of local 
military power. 

By 362 b.c. peace had been restored and both Datames and Ariobarzanes were 
killed, although the son of Datames, Sysinas, received his father's satrapy to rule while 
Aroandas received much of the Aegean coast. Mausolos and others were forgiven. 
The Achaemenid Empire had not collapsed thanks to the inability of the rebels to 
work together. Artaxerxes II was murdered in 359 and succeeded by Ochus, one of 
his sons, who succeeded in killing most of his near relatives and taking the throne 
name Artaxerxes III. 

The new king was a cruel but strong ruler, and he resolved to curb the power of the 
satraps, one of which, Artabazos, an Achaemenid prince, had been satrap at Daskylion 
in Phrygia. He and Aroandas in Mysia refused to disband their Greek mercenaries at 
the order of the king and revolted. The newly appointed satrap in Phrygia, 
Tithraustes, together with Autophradates of Sardis and Mausolos were ordered by the 
king to crush the rebels, but Athens supported the rebels who initially were successful 
against the royal forces. Then Artaxerxes threatened Athens with war, and Athens 
withdrew support of the rebels. Thebes took the place of Athens in support of 
Artabazos, but after a few successful encounters this coalition disintegrated and 
Artabazos fled to Philip of Macedonia about 353 B.C., while Aroandas remained a 
rebel. Nonetheless, most of Anatolia returned to obedience to the Achaemenids, 
enabling Artaxerxes to turn his attention to Egypt. It seems that he earlier had 
successfully suppressed a revolt by the Cadusians and possibly other Iranian peoples, 
since they are found later in his army (Diodorus XVII, 6, 1). 

On the way to Egypt the royal army had to reconquer Phoenicia which had been 
free from Achaemenid rule, but the campaign in Egypt was more difficult than 
foreseen. The chronology is uncertain, but after much fighting the Persians had to 

The standard account of the Anatolian "Kleinasien zwischen Agonie des Perserreiches und 

revolts by W. Judeich, Kleinasiatische Studien hellenistischem FrUhling," Anzeiger der OAW, 

(Marburg, 1892), 190-297, is still the most 112 (1975), 274-82, who discusses a trilingual 

etailed account, to be corrected by Heichelheim, inscription from Xanthos. 
°P- tit. [n. 155], 35^t0. See also M. Mayrhofer, 



132 Chapter V 

retreat from Egypt about 350 B.C. which led to a new revolt in Phoenicia. By 345 
Phoenicia was reconquered and placed under the satrap governing Cilicia Mazaios or 
Mazdai, as he is called on his coins with Aramaic legends. 162 The reconquest of Egypt 
was led by the king himself, and his army had many Greek mercenaries from cities 
other than Athens and Sparta, and they fought against other Greek mercenaries on the 
Egyptian side. The Achaemenid mercenaries won in 343 B.C., and the Egyptian 
leaders fled to the south to Upper Egypt. Egypt had returned to Persian rule, but it 
was not entirely subdued, while the quick capitulation of the land to Alexander the 
Great indicates the attitude of the Egyptians towards Achaemenid rule. 

In Greece, many had preached urging Greek unity against the Persians, the most 
famous of whom was the Athenian Isocrates (between 436-338 B.C.). In his 
panegyrics (esp. 161-62) he urged the Greeks to unite in a pan-Hellenic union, but 
many preferred Persian friendship and gold to cooperation with their neighbors, and 
it was left to the Macedonians to enforce the unity. The Achaemenid Empire was ripe 
for the plucking in many respects, political, economic, social and even religious, to 
which questions we may briefly turn. 

It must be emphasized again that the Iranian satrapies of the empire cannot be 
compared with Babylonia and Egypt, for Achaemenid attempts to make a unified 
system of weights and measures, roads and postal system, currency, taxation, etc. were 
not successful everywhere, and each satrapy had an independence in these matters. A 
unified coinage, for example, did not exist, as we see from the proliferation of satrapal 
issues in the fourth century, all in the western part of the empire. The east, it seems, 
had no coinage, and we may suppose that the pastoral and agricultural societies 
continued to exist much as ever, with little connection with events on the Aegean and 
eastern Mediterranean coasts. Revolts were surely frequent even in the east, but there is 
no evidence of risings to regain liberty by an Iranian ethnic group or nation, such as 
the movements for independence of Egyptians, Phoenicians or Ionians in the west. 
From the booty secured by Alexander, it seems clear that the Achaemenid courts 
became ever growing treasury houses of precious metals and luxuries from all over 
the empire. The gold coins or darks of the royal house were used to hire mercenaries, 
or for bribery. Unfortunately, we have no sources to report the effects of the 
hoarding of gold and precious objects by the rulers on the economy of the empire, but 
it hardly can have been propitious. The loss of tax revenues from Anatolia, the Syrian 
coast and Egypt, as well as India, must have hurt the royal treasury and caused a 
heavier load on the provinces still loyal to the Achaemenids. 163 In Babylonia, the 
only province where only sparse information about prices of goods, slaves, etc. can be 
found, one can venture a surmise that the existing gap between the Achaemenid 
aristocracy and the subject population grew more and more in the fourth century as 



162 On the Phoenician resistance see J W. from the Hindukush region to the east had no 
Betlyon, "A New Chronology for the Pre- Achaemenid satraps, and no evidence exists that 
Alexandrine Coinage of Sidon," ANSMN, 21 they had been part of the empire for years before 
(1976), 28-35, with references. his invasion. Cf. F. Altheim and R. Stiehl, Die 

163 The date of the loss of the satrapies of the far Aramaische Sprache miter den Achaimeniden (Frank- 
east (India, Sattagydia, Gedrosia) cannot be deter- furt/M, 1963), 127. 

mined, but by the time of Alexander all of the areas 



Achaemenids J 33 

compared with the earlier years of the empire. 164 Just how far the 'feudal' lands of the 
great Achaemenid aristocracy grew in the various parts of the empire is impossible to 
determine, but scattered notices in Classical sources give the impression that such land 
holdings increased rather than decreased. With rebellions and losses in wars in the 
fourth century the economy of the empire was hardly prosperous. On the other hand, 
the growing number of Iranian names in high positions all over the empire, and in 
lower positions, too, imply a spread of control everywhere of Iranians over the local 
populations. 165 Yet this bureaucratic domination does not seem to have greatly 
influenced the life of the subject peoples, since after Alexander vestiges of Persian 
culture and influence in non-Iranian areas do not appear in sharp relief. What seems 
certain is that already before the coming of Alexander, Greek influence was much 
stronger than Iranian in the western part of the empire, while Greek coins apparently 
were accepted even in the easternmost provinces. 166 

In religions, the oft mentioned change in the religion of the Achaemenids from the 
time of Darius to Artaxerxes II, on the basis of the appearance of Mithra and Anahita 
in the royal inscriptions, may be more apparent than real. On the other hand, to 
suppose that there was no change in the religion of the Persians throughout the life of 
the empire is also exaggerated, for the religious outlooks of the Greeks, Hebrews, 
Babylonians and others changed, hence no development on the part of the 
Achaemenids is most unlikely. The Zeitgeist of the fourth century B.C., one would 
imagine, favored syncretism, or at least mutual influences between religions, and we 
may postulate a greater reciprocity of influences in the Near East at that time than 
earlier. The influence which Zoroastrianism had on Judaism cannot be proved in 
detail, but the parallels are so striking in the conception of angels, dualistic ideas and 
eschatology that an independent development is most unlikely. 167 Likewise the 
erection of statues to Anahita is surely an Iranian borrowing from the religions of the 
ancient Near East. Questioning of the authority of ancient deities was in the air 
everywhere in the fourth century and probably the divinely sanctioned power and 
authority of the great king were likewise in doubt, while the background for the 
flourishing of mystery and savior cults of the Hellenistic age was already prepared by 
the end of the Achaemenid period. Much has been written about a 'Zoroastrian' 
calendar which many scholars believe was introduced into Iran by Artaxerxes I in 
441 b.c. based on an Egyptian model, but this date rests on conjecture. Aramaic 
documents from various parts of the empire indicate that the Babylonian calendar 
was used throughout the existence of the empire and the use of different names for the 
months does not prove the introduction of a new calendar. 168 The use of the calendar 
to prove a change in the religion of the Achaemenid under Artaxerxes I is not valid 

16 * Cf. the summary in J. Oelsner, "Krisener- l66 See Schlumberger, op. cit. [n. 112], 6-30. 

scheinungen im Achaimenidenreich im 5. und 4. l67 Cf. R. Frye, "Qumran and Iran," in J. 

Jahrhundert v.u.Z. ," in E. C. Welskopf, Hellenische Neusner, ed., Christianityjudaism and Other Greco- 

lis 2 (Berlin - 1974 ). 1049-63. Roman Cults, part 3 (Leiden, 1975), 167-73, with 

See the names in W. Eilers, Iranische further references. 
Beamtennamen in der keilschriftlichen Oberlieferung 168 E. J. Bickerman, "The Zoroastrian Calen- 

(Leipzig, 1940), and for Asia Minor A. Goetze, dar," AO, 35 (1967), 204-05, with further 

Kulturgeschichte des Alten Orients: Kleinasien references. 
(Mtinchen, 1957), 210-12 



134 Chapter V 

and should be discarded. The development of cults and various practices, including 
syncretistic beliefs under the later Achaemenids, on the other hand, is plausible even 
though the evidence so far is lacking. Probably the Zoroastrianism which we know 
from the Sasanian and later periods was the religion of the majority of Iranians even at 
that time, though not necessarily the faith of the members of the Achaemenid royal 
family, nor the popular religion of many villagers in different parts of Iran. The 
supposition of abrupt or contradictory changes in Zoroastrianism of Achaemenid 
times, based on an enigmatic word or reference here and there, is hardly warranted, 
and it is easier to assume continuity, albeit with modifications and adoption of new 
rituals and practices, than fundamental changes. Obviously the spread of practices, 
such as the exposure of the dead to vultures, took much time before it became a 
religious ordinance, the breaking of which was a grave sin. To brand a burial from 
Achaemenid times as non-Zoroastrian involves an error of imposing 'orthodox' 
practices of later Zoroastrianism on earlier periods, and should not be- assumed 
automatically. The converse, however, evidence for the exposure of bodies, would 
lead an archaeologist to declare that the deceased were Zoroastrians, since we have no 
evidence that non-Zoroastrians practiced this in the Iranian world. 

Tied somewhat to religion is the question whether allegiance to the house of the 
Achaemenids weakened towards the end of the empire. We have seen the growth of 
serious revolts against central authority, the spread of Greek culture and ideas, and the 
excesses of both the court and the Persian aristocracy. All must have combined to 
reduce the influence or even legitimacy of the great kings in the eyes of many of their 
subjects, including Iranians, witness the many revolts of the Cadusians. Possibly of 
great significance was the usurpation of power by a eunuch Bagoas who had his 
master Artaxerxes HI Ochus poisoned, and then placed one of his sons Arses on the 
throne in 338 B.C. But Arses tried to wrest control of affairs from Bagoas and lost his 
life to the poison of the eunuch after less than two years of rule. So in 336 B.C. the last 
Achaemenid prince in direct line had perished without leaving a son. Bagoas, unable 
to become king himself, looked around for an Achaemenid whom he might easily 
control, and he selected a grandson of Ostanes, brother of Artaxerxes II, known in 
Greek sources as Codomannus or Darius III. Shortly afterwards Bagoas was killed by 
Darius, but the harm to royal prestige already had been carried out by Bagoas; and 
even though the last Darius at first showed himself to be capable, when Alexander 
appeared on the scene, the fame and glory of the Achaemenids could not be restored. 
Darius did reconquer Egypt, however, which had broken away again from 
Achaemenid rule after the death of Artaxerxes III, but its submission to Alexander by 
the satrap Mazakes without resistance indicates the tenuous hold the Persians had over 
Egypt. 

Meanwhile in Greece, Philip of Macedonia had defeated the Greek city states at 
Chaeronea, the same year that Artaxerxes III was murdered, and he took over the 
leadership of a united Greek crusade against Persia. The intrigues of Philip, Athens, 
Achaemenid satraps and the great king are part of Greek history. The murder of 
Philip in 336 B.C. before the re-conquest of Egypt by Achaemenid forces brought a 
lull in diplomatic activities, and undoubtedly Darius at first did not consider 
Alexander a serious threat, since he did not actively support Athens in her search for 
aid in an uprising against the Macedonians. Demosthenes tried to persuade the 



Achaemenids 135 

Athenians that Alexander represented a greater threat to Greek liberties than Persia, 
but he failed, especially after Alexander crushed the Greek revolt and destroyed the 
city of Thebes. The Macedonians crossed the Hellespont and invaded Asia and the end 
of the empire was in sight. 

Perhaps here one may speak briefly of the legacy of the Achaemenids in the lands 
which they occupied for such a long period. From the Elamite documents and 
Aramaic papyri from Egypt we gain some insights into a huge empire which did try 
to unite all the peoples in it with some common features, one of which was law. The 
concept of a universal 'king's law' over the local 'laws' of subject peoples was a legacy 
which the Romans followed and which, as far as we know, was an innovative feature 
of Achaemenid rule. 169 It is obvious, of course, that the Achaemenids borrowed 
heavily from the ancient Near East, especially Babylonia, in constructing their 
institutions and bureaucracy ; for example, in the domain of law the parallels between 
the Babylonian ardu 'slave' or 'menial' and Old Persian bandaka, both juridical persons 
who could own property and occupy high posts, is striking. The 'empire' and the 
'great king' impressed the Greeks even though they were enemies. This did not 
change the Greek esteem for themselves as opposed to the 'barbarians,' and the Persian 
continuation of earlier despotism could not have found favor in Greek eyes. The Jews, 
on the other hand, fared well under the Achaemenids, which speaks for some cultural 
or religious affinity of the two peoples. 170 The lack of sources is especially frustrating 
for the dark period of the last century of Achaemenid rule, and there is little hope 
of new discoveries of source materials. In art, too, the clues to the meaning of 
Achaemenid creations must be sought in symbols, many undoubtedly age-old Near 
Eastern symbols which had lost their meaning by Achaemenid times. In any case, the 
art was never personal; it was not so much a certain king who is portrayed on stone 
but 'kingship.' The anthropocentrism of Western art is not found here. Likewise there 
is an absence of religious art in Iran, where decoration is more important than 
function. So form and symbolism are characteristics of Achaemenid art, and by the 
time of the Sasanians, the heirs of the Achaemenids, the symbolism for the most part 
had been reduced to forms, the forerunner of Islamic art. 

It is perhaps in the memory of the great oecumene - the one world of the 
Achaemenids - that the greatest legacy is found, for Alexander, who usurped the role 
of 'world hero' from Cyrus and his successors, nonetheless was following in their 
footsteps, and that conception was passed to posterity by Alexander, perhaps the real 
heir of the Achaemenids. 

,69 Josephus in his Contra Apion (II, 270-71) tion in Juda zur Achamenidenzeit (Berlin, 1973) and 

says that Apollonius had a high regard for the laws P. P. Backroyd, Israel under Babylon and Persia 

of the Persians. (Oxford, 1970). 
See H. Kreissig, Die sozialskonomische Situa- 



CHAPTER VI 
ALEXANDER THE GREAT AND THE SELEUCIDS 



Literature: The literary sources are all Greek or Latin, since this period of the history of Iran would be 
almost a total blank if we depended on other sources. Discussion of the lost histories of Alexander by 
Ptolemy, Aristobulus and Clitarchus is important but that is the task of a Classicist, while the surviving 
works of Arrian, Quintus Curtius, Plutarch, Diodorus and Justin-Trogus may be combed for occasional 
details about Iranian affairs. The coinage of Alexander is well summarized by A. R. Bellinger, Essays on the 
Coinage of Alexander the Great (New York, 1963). Other than this literature, for Iran we have the 'folk 
literature' of the Alexander Romance in the Shahname and in many Arabic, Armenian, Syriac and Persian 
works from the Islamic period. These show the change in the view of Alexander by Iranians from a hated 
conqueror to an Iranian hero who was a scion of the Achaemenid family by a secret marriage between 
Philip of Macedonia and an Achaemenid princess. There is a vast literature on the Alexander romance, 
but for its Oriental versions consult the old but still useful F. Spiegel, Die Alexandersage bei den Orientalen 
(Leipzig, 1851), 72 pp., and J. A. Boyle, "The Alexander Legend in Central Asia," Folklore, 85 (1974), 
21 7-28, for further references. The change in Alexander from a negative, enemy figure to an Iranian hero 
is discussed by A. Abel in "La figure d'Alexandre en Iran," La Persia e it mondo Greco-Romano, Acad, dei 
Lincei, Quaderno 76 (Rome, 1966), 119-36. Whether the Sasanians were those responsible for again 
changing the figure of Alexander from a positive to a negative personality is difficult to determine, but 
more likely it was a gradual change over centuries. The Romance is of significance here only to show the 
Iranian attitudes towards Alexander, and has little or no historical value. 

The geographical literature in Greek and Latin, which gives more than mere geography, is especially 
important for this period of history, not only Strabo, Pliny, and Ptolemy (for the last of which see I. 
Ronca, Ptolemaeus Geographia, 6, 9-21, Ostiran und Zentralasien [Rome, 1971]), but also fragments of 
geographers found first in C. Mtlller, Geographi Graeci Minores, 2 vols. (Paris, 1855), then in Frg. Hist., 
and A. Riese, Geographi Latini Minores (Heilbronn, 1878). The Lexicon of Hesychius in the edition of K. 
Latte in 3 vols. (Copenhagen, 1953-79) is useful for names, as is the Ethnika of Stephan of Byzantium, ed. 
by A. Meineke (Berlin, 1879). For the geography of the Iranian plateau or Central Asia there is nothing 
comparable to the study on northern Iraq by L. Dillemann, Haute Mtsopotamie Orientate el pays adjacents 
(Paris, 1962), or even H. Hiibschmann, Die altarmenischen Ortsnamen (Strasbourg, 1904). The Tabula 
Peutingeriana, a medieval map, ed. K. Miller, Itineraria Romana (Stuttgart, 1916) and the Ravenna 
cosmography, ed. J. Schnetz, Itineraria Romana (Leipzig, 1940), with his translation (Uppsala, 1951), both 
give geographical names from an earlier period of history, although identifications of place names are 
frequently most difficult. The work of J. Marquart (Markwart) is found in his bibliography by H. H. 
Schaeder in Markwart, Wehrot und Arang (Leiden, 1938), *53-61, while the older study by W. 
Tomaschek, "Zur historischen Topographie von Persien," Sb WAW, 102 (1883) is still very useful for 
geographical identifications in Iran and Central Asia. 

Greek inscriptions from the Iranian plateau are collected in the Supplementum Epigraphicum Graecum, 
ed - b yJJ- E. Hondius and A. G. Woodhead (Leiden, 1 923-) ; in the Revue des Etudes Grecques and Hellenica 
both by L. Robert, adding to the old Orientis Graeci Inscriptiones Selectae, by W. Dittenberger (Leipzig, 
1903). The few Aramaic or heterographic inscriptions may be found in R. Degen und W. MUller, Neue 
Ephemerisftir Semitische Epigraphik (Wiesbaden, 1972-) with extensive bibliographies. 

Art and archaeology titles may be found in L. Vanden Berghe, ed., Bibliographie analytique de 
larchhiogie de I'lran ancien (Leiden, 1979), 256-70. The two extensive archaeological excavations of 
interest for this period are Ay Khanum in northern Afghanistan and Seleucia on the Tigris. For the former 
we have P. Bernard, Fouilles d'Ai Khanoum, MDAFA, 21, 2 vols. (Paris, 1973), and many articles on the 
progress of the excavations in yearly reports of CRAI. Seleucia was excavated before World War II by L. 
Waterman, and we have a series of reports in the University of Michigan Studies, Humanistic Series, vol. 
32 by N. Debevoise on Parthian Pottery, vol. 36, on Stamped and Inscribed Objects by R. H. McDowell, vol. 
37, on Coins by McDowell, and vol. 45 Clay Figurines, by W. van Ingen (Ann Arbor, 1939). More 
recently we have Clark Hopkins, ed., Topography and Architecture of Seleucia on the Tigris (Ann Arbor, 
973). The excavations of Dura Europos on the Euphrates are mostly concerned with the Parthian period 



i^v Chapter VI 

and do not help us much for earlier times. General works are L'Orient HelUnisi by D. Schlumberger 
(Paris, 1970) for art, and E. Will, Hisloire politique du monde hellinistique, 2 vols. (Nancy, 1967) which 
supersedes the older works of E. R. Bevan, The House of Seleucus, 2 vols. (London, 1902) and A. Bouche- 
Leclercq, Hisloire des Seleucides, 2 vols. (Paris, 1914). Will speaks of the technical problem of ruling the 
vast domain of the Seleucids with relatively few Greeks and Macedonians, and a 'human' problem of the 
co-existence of the newcomers with the established population, and what one might call two groups 
wielding local authority sometimes in cooperation, at other times in opposition. The book by S. K. Eddy, 
The King is Dead (Lincoln, Neb., 1 961 ) investigates evidence for local opposition to Hellenic imperialism 
and concludes it was almost wholly a religious resistance, though this oversimplification is dubious. The 
classic work of E Bikerman, Institutions des Seleucides (Paris, 1938), is still the best on the subject, but 
internal affairs in the east are hardly mentioned in it. The two syllabi, one for the Sorbonne by A. 
Aymard, Lesgrandes monarchies hellenistiques en Asie aprts la mort de Seleucos I er (Paris, 1 965), 21 2 pp., and 
the other for Yale by C. Bradford Welles, The Hellenistic World, A History (New Haven, 1961), 142 pp., 
are both interesting for students to read but are little more than summaries. 

Since only archaeology provides any new material for this time period, the book of D. Schlumberger is 
especially important for new vistas in material culture. The excavation of Ay Khanum has revealed the 
strength of Greek institutions, language and culture, in the far-flung city foundations of Alexander and the 
Seleucids, and Schlumberger rightly underlines the strong persistence of Hellenism in the distant Orient 
long after the disappearance of Greek rule. The small seaport of Failaka, or ancient Ikaros, in the Persian 
Gulf off the coast of Kuwait, also provides evidence of the extensive Greek influence on architecture in 
Seleucid times. Cf. K. Jeppesen, "A Royal Message to Ikaros" in Kuml, 9 (Aarhus, 1960), 153-98. 

The site of Khalchayan, on the right bank of the Surkhan Darya near present Denau in the Uzbek SSR, 
although extending into later periods, also gives some architectural information, supplementing Ay 
Khanum, on Seleucid rule in Central Asia. Cf. G. A. Pugachenkova, Khalchayan (Tashkent, 1966) and her 
Skulptura Khalchayana (Moscow, 1971). The erratic writings of Franz Altheim, while containing many 
vignettes of interest and value, must be used with extreme caution. See his bibliography, ed. by E. Merkel 
(Frankfurt/M, 1958), and with additions in his Festschrift, Beitrage zur Allen Geschichte und deren 
Nachleben, ed. by R. Stiehl, 1 (Berlin, 1969). Obviously, various articles on the results of excavations, on 
coinage, or other special subjects provide additions to our knowledge of the Seleucids in Iran and Central 
Asia, but little change need be made in the scheme of coinage made by E. T. Newell, The Coinage of the 
Eastern Seleucid Mints (New York, 1938). An Inventory of Creek Coin Hoards (New York, 1973) by M. 
Thompson, O Morkholm and C. Kraay, is not as valuable for Seleucid history as for other Hellenistic 
kingdoms, but one should note that the concentration of Greek settlements on major trade routes is clear 
from the hoard finds. On the cities founded by the Seleucids, the work by V. Tscherikower, "Die 
hellenistischen StadtegrUndungen von Alexander dem GroBen bis auf die ROmerzeit," Philologus, 
Supplementband XIX, Heft 1 (Leipzig, 1927) is still the classic work, but add the extensive discussion on 
cities in eastern Iran and Central Asia by G. A. Koshelenko, Crecheskii Polis na ellinisticheskom Vostoke 
(Moscow, 1971), 113-60, with texts of inscriptions found in the east. In the sources, much confusion 
exists about the identifications of the various Alexandrias, not all of which were founded by Alexander 
the Great, for some later rulers used his name just as in the case of coinage. For Alexander's policy towards 
the Persians, the articles by H. Berve, "Die Verschmelzungspolitik Alexanders des GroBen," Klio, 31 
(1938), 135-68, supplemented by E. Badian's "The Administration of the Empire," Greece and Rome, 12 
(Oxford, 1965), 166-82, provide the best analysis. 

ALEXANDER IN IRAN AND CENTRAL ASIA 

The conquests of Alexander are the best known campaigns of ancient history and the 
enormous literature about him and his deeds makes any account of his actions 
redundant here. 1 We shall be concerned here only with the impact of Alexander on 

1 An excellent and detailed survey of both (Darmstadt, 1972) [a survey of literature about 

sources and secondary works on Alexander is Alexander]; see also H. Berve, Das Alexanderreich 

provided by E. Badian in "Alexander the Great, auf prosopographischer Grundlage, 1 (Munich, 

1948-67," The Classical World, 65 (1971), 37-83, 1926), 253-76, a most convenient reference work 

plus his "Some Recent Interpretations of Alex- on the Greco-Macedonians and Iranians active in 

ander," Entretiens sur I'antiquite classique, 22 (Gene- the time of Alexander, 
va, 1976), 279-303; J Seibert, Alexander d. Gr. 



Alexander the Great and the Seleucids J 39 

Iran and Central Asia, and at the onset we may ask two questions : how did Alexander 
treat the Iranians who comprised the satraps and upper ruling group of the 
Achaemenid Empire and how did the Iranians regard Alexander and his successors? 
The last may be viewed through the sparse and scattered bits of evidence we find in 
our sources, or in the eastern versions of the 'Alexander Romance.' 2 Obviously no 
analysis of the Classical sources, nor of Alexander's campaigns, of his character or 
dreams, can be made here and only the Iranian views need be considered for this 
volume. 

When we remember that Iranian, and especially Persian, nobles dominated the 
government of the Achaemenid empire, we may ask what happened to them ? In Asia 
Minor all of the Iranian satraps were replaced by Macedonians, or by native dynasts 
such as the princess Ada in Caria, and the same policy was followed in Syria and 
Egypt. When Alexander came to the east after the battle of Gaugamela, however, his 
policy changed, and he re-appointed Mazaios, the important satrap of Syria and 
Mesopotamia under Darius III, over Babylonia. A number of questions arise, 
however, which are not easy to answer: first whether Mazaios was really the 
Achaemenid satrap of the 'Fertile Crescent,' or most of it, at the time of Alexander's 
conquest, and a second question, why Alexander appointed him satrap of Babylonia. 

Mazaios was the ruler of Syria and northern Mesopotamia at the time of 
Alexander's invasion, whereas his earlier post as satrap of Cilicia under Artaxerxes III 
had already established his authority to issue silver coins in his own name. We may 
conjecture that Mazaios took advantage of the weakness of the empire under the last 
Darius to consolidate and expand his own power and influence. In any case, Mazaios 
as the satrap of the northern part of the 'Fertile Crescent' had to bear the brunt of 
fighting against Alexander between the battles of Issus and Gaugamela, and at the 
latter battle he commanded the right wing and almost defeated Alexander. 3 Mazaios 
was certainly an important Persian noble, but it was probably his ability and bravery, 
as well as the believable report in Curtius (IV, 16) that Mazaios after the battle fled to 
Babylon with his troops and took charge of the city, which enhanced his position. 
When Alexander came, Mazaios surrendered to him and was named satrap of 
Babylonia (Arrian III, 16, 4; Curtius V, 1, 44). That he was allowed to issue similar 
coinage to the series he had struck while governor of Cilicia and later Syria has been 
proposed, with the reason for this unusual procedure being an agreement between 
Mazaios and Alexander involving the surrender of Babylon to the latter. 4 This 
suggestion seems a more likely reason for Mazaios' right to continue his striking of 
satrapal coins than Alexander's respect for a brave enemy or any other sentimental 
reason. The numismatic evidence for these coins having been struck in Babylonia 
after Gaugamela, however, is most uncertain and until numismatists uncover hoards 

Some Iranian sources are specific in their anti- trian tradition may be overly tendentious in this 

Alexander statements, that he burned the Avesta, regard. 

Killed some of the priests and learned men of Iran, 3 E. W. Marsden, The Campaign of Gaugamela 

extinguished fires, and was a great destroyer. For (Liverpool, 1964) is a military analysis with little 

* e references see M. Boyce, Zoroastrians [ch. 5, n. concern for politics or anything else. 
128], 78, (where the 'Frataraka' temple at Persepo- 4 Cf. A. R. Bellinger, Essays on the Coinage of 

hs was built after Alexander, not before his Alexander the Great, ANS, Numismatic Studies, 1 1 

invasion). Whether 'mass slaughters' of priests took (N.Y., 1963), 61-64, where other literature is 

P'ace, however, is highly dubious and the Zoroas- given. 



140 Chapter VI 

and more firm evidence the coinage should not be used to substantiate a presumed 
policy of Alexander in granting such a right of striking coins to subordinates. In any 
case, Mazaios was the first noble Persian to be appointed by Alexander to an 
important post. His power was circumscribed, however, by the appointment of two 
Macedonians, one in charge of the army and the other to collect taxes (Arrian HI, 16, 
4; Curtius V, 1, 43; Diodorus XVII, 64, 5). Nonetheless, with this move Alexander 
had appointed a prominent official of the old government to a high post, and whether 
the conqueror acted out of a sense of desire for conciliation or the need to have an 
experienced local person, whose wife may have been a Babylonian to judge by the 
names of his two sons, in charge of local government cannot be determined. We also 
cannot decide the identity of the person who may have struck coins in the time of 
Alexander somewhere east of the Euphrates with Aramaic legends bearing his name 
mzdk, identified by some as the Achaemenid satrap of Egypt who surrendered to 
Alexander and then joined his court. 5 These coins, with the Attic owl on the reverse, 
are so enigmatic, however, that they cannot be used in any historical identifications. 
These coins may be indications, however, of Alexander's pragmatism in not 
instituting quick and momentous changes in the coinage of the lands he conquered. 
Since the rulers of the satrapies for generations had been Persians, Alexander perforce 
had to turn to Persians with experience in rule rather than to Babylonians or others. 
Furthermore, since Babylonia was regarded by the Achaemenids as a place of winter 
residence while their homeland, the Iranian plateau, was more suited for summer, 
Alexander may have considered the battle of Gaugamela the key to the whole east, of 
which Babylonia was the most populated and richest part of the 'Iranian part' of the 
empire, and therefore his policy towards the conquered may have been revised. 

This point of view brings us to the oft-disputed theory that Alexander preached the 
idea of the 'unity of mankind,' in his conquests, a kind of leitmotif for his 'crusade.' In 
spite of the writings of W. W. Tarn and others, there is more evidence for a proposed 
Iranian-Macedonian partnership in rule than for any universal 'equality' or policy of 
'conciliation' directed towards all conquered peoples. 6 It is hardly possible, of course, 
to enter the mind of Alexander, but he probably realized that some sort of 
cooperation with the Iranians was needed to secure and maintain his rule in the east. A 
survey of the satraps appointed by Alexander supports this contention. After 
Babylon, Alexander confirmed the satrap of Susa, Aboulite's, in his post, since he had 
surrendered the treasures of the city to Alexander, but here too he left several of his 
own officers as commander of the garrison, collector of taxes, and others (Curtius V, 
2, 16). The ruler of the land of the Uxii (read: Huzi) was a relative of Darius called 
Madates (Diod. XVII, 67, 4; Curtius V, 3, 4) who resisted Alexander in the 
mountains and presumably was either killed or at least removed from his rule of the 
province. The satrap of Persis, Ariobarzanes, also offered strong resistance to the 
Macedonians, but after defeat he fled to the north and eventually with his father 
Artabazos and his brothers surrendered to Alexander, who, however, did not reward 
him with another position. In Persis, after the burning of Persepolis, Alexander 

5 Ibid., 66, with further references. Alexander set the stage for the later ideas of world 

6 W. W. Tarn, Alexander, \ (Cambridge, 1948), brotherhood or unity supposedly preached by 
146 foil. Few scholars today would maintain that Zeno and the Stoics. 



Alexander the Great and the Seleucids 141 

appointed a Persian Phrasaortes as satrap apparently with the usual controls and 
checks of Macedonian or Greek officers at his side to collect taxes and insure order. 
Thus the pattern is repeated, of keeping the Iranian governor as a figurehead with the 
real power in the hands of the conquerors. At Persepolis, the Persian in command of 
the site, Tiridates, was also reinstated in his position with a Macedonian garrison 
(Curtius V, 6, 11). According to Arrian (III, 19, 2), Oxathres, son of Aboulites, was 
made satrap of Paraitakene, the land north of Persis, the chief city of which was Gabai 
or Isfahan. In Media Alexander appointed a certain Oxydates as satrap under the same 
conditions as other satraps, and in Parthia and Hyrcania (hodie Gurgan) he named 
Amminapes, a Parthian who had accompanied Alexander on his expedition from an 
exile in Macedonia. Shortly after the former satrap of Parthia and Hyrcania, 
Phrataphernes, surrendered to Alexander, after the death of his king Darius III, 
Alexander reinstated him in his old post and removed Amminapes. Another satrap, 
Autophradates, was also confirmed in his rule of the Caspian Sea province of the 
Tapuri (medieval Tabaristan). Satibarzanes, satrap of Aria (Herat) had assisted Bassos, 
the satrap of Bactria and Sogdiana in the murder of Darius III, but had then 
surrendered to Alexander who confirmed him in his satrapy. But this satrap revolted 
and after much fighting was killed. This was the beginning of a resistance to 
Alexander by the satraps of the east who did not submit but opposed the conqueror, 
and since Darius was dead and Alexander had taken his place, the satraps who did not 
submit probably were considered by many as rebels to their rightful ruler. 

Alexander's policy of placing an Iranian as satrap over a conquered province with 
Greco-Macedonian generals and troops beside him was continued, on the whole, also 
in the farther east, although we find a Macedonian MenOn made satrap of Arachosia 
after the Iranian satrap Barzaentgs, one of the murderers of Darius III, had fled to 
India. In the Hindukush region to the north, an Iranian called Proexes by Arrian (III, 
28, 4) was made the satrap and in Bactria another, Artabazos, in place of Bessos, who 
had proclaimed himself Artaxerxes IV, successor of Darius, and had become the 
leader of the opposition to Alexander in the east. In the border areas of Sogdiana in the 
north and India to the east Alexander did not find satraps but rather principalities 
which may have owed some allegiance to the Achaemenids, but which were really 
independent. In these areas Alexander had to fight more than elsewhere, not because 
of an Iranian 'national revival' as has been claimed but because of virtual independence 
from Achaemenid rule, the local rulers unwilling to submit to a new centralized 
authority. Both on the Central Asian and the Indian borderlands of the Achaemenid 
Empire little more than a nominal allegiance had been given the Achaemenids. This is 
indicated by the earlier incorporation of Sogdiana, which was virtually independent, 
into the satrapy of Bactria, and by the difficult fighting which Alexander had to do in 
this part of the world. The subsequent revolts of the Sogdians and Bactrians after they 
had been subdued by Alexander put a severe strain on the Macedonian army (Arrian 
IV, 1, 5; Curtius VII, 1, 14). It was in Bactria and Sogdiana that Alexander had to 
establish strong garrisons to maintain his rule, but strangely he did not make a 
Macedonian governor, rather granting the combined satrapy to Artabazos in 330. At 
the latter's request, because of old age, Alexander in the winter of 328 did replace him 
with a Macedonian Amyntas, and large contingents of troops had to be left under 
Amyntas to maintain Macedonian rule, while Alexander marched to India (Arrian 



142 Chapter VI 

IV, 22-3). It is highly probable that more recruits, or mercenaries as Curtius (VII, 10, 
11) calls them, had to be brought from the west to replace losses and to provide 
needed garrison troops. Some of the replacements came unwillingly, if we are to 
believe the report that the garrison of Bactra revolted on hearing a rumor of the death 
of Alexander in India and unsuccessfully sought to return to their European 
homelands (Curtius IX, 7, 2). Thus, a new pattern emerges of relying on Iranian 
officials to run the provinces of the Achaemenid Empire which were secure under 
central authority, whereas the frontier areas in the east, he soon discovered, needed a 
strong new administration of Macedonian officers and troops to insure continued 
allegiance to Alexander. Only a strong presence of Greco-Macedonians could guard 
those frontiers against nomadic enemies, as well as hold down the local population 
who had held little or no loyalty to the old Achaemenid government, and 
consequently showed the same attitude towards the heirs of the Achaemenids. For 
only military force could create a new allegiance among the settled folk to the new 
rulers against the nomads of the steppes. 

The growing 'Orientalization' of Alexander, well described by Arrian (VII, 6, 2), 
was really a 'Persianization, or an Achaemenidization,' at least in regard to matters 
such as clothes, customs and various practices, such as the oft-disputed proskynesis. 
The latter, it is now recognized, had nothing to do with the divinization of Alexander 
which was a Greco-Macedonian problem and not an Iranian one. 7 The prostration of 
menials, or the raising of the hand to the mouth on the part of nobility before the 
ruler was customary among the Achaemenids and elsewhere, and the Greco- 
Macedonians did not change this practice among the Iranians no matter how much 
disputed it may have been for the Westerners. Alexander's attempts at conciliation 
with the ruling Persians undoubtedly met a favorable response among many who 
retained their positions, but the majority of Iranians, we may suppose, regarded 
Alexander as a usurper. The policy of intermarriage between Greco-Macedonian 
officers and Iranian noblewomen culminating in a great banquet at Susa (Plutarch, 
Alex., 70; Arrian VII, 4, 4) may have been part of the hope of Alexander that a 
mixture between Greco-Macedonians and Iranians would produce a progeny able to 
rule the vast empire. From the sources, it seems that Alexander went out of his way in 
favoring the Iranians so as to bring them to partnership in rule. The selection of many 
noble Iranians as his ovyyevels 'kinsmen' angered his Macedonian generals and 
companions, who resented the advancement of the Iranian nobles to equality with 
them (Arrian VII, 11,6), while several Iranian contingents in the army received a 
status equal to the eralpoi or 'companions' of the king (Arrian VII, 6, 4) at least in the 
eyes of some Macedonians. 8 Much has been written about the intentions of 
Alexander, and the eventual mixture mentioned above seems to have been his 
intention, but during his lifetime a dual status of two ruling peoples appears more 
likely, for obviously he needed the existing bureaucracy to run the empire, and the 
seeds of a kind of 'double rule' of the Seleucids, best exemplified in the double us? of 

1 On proskynesis see J. P. V. D. Balsdon, "The 8 On the resentment and mutiny of the 

'Divinity' of Alexander," Hisloria, 1 (1950), 380- Macedonians against the advancement of Iranians 

82, and Feodora von Sachsen-Meiningen, "Pros- in Alexander's service, see E. Badian, "Alexander 

kynesis in Iran," in Altheim, Hunnen, 2 [ch. 5, n. the Great and the Unity of Mankind," Historia, 7 

68], 126-27, and 152-54. (1958), 428. 



Alexander the Great and the Seleucids 143 

the Greek and Aramaic languages, were laid by Alexander. Iranian satraps were 
reinstated as heads of the provincial bureaucracies while the military power was in the 
hands of Alexander's officers. The conquered eventually had to be integrated into the 
army formations with Macedonian weapons, while the practices and customs of the 
conquered had to be respected by the foreign, Greco-Macedonian soldiery. Thus, we 
hear of the adoption of Iranian dress and practices by the conquerors, but it was the 
other way in military matters such as the adoption of Macedonian spears by units of 
Iranian soldiers in Alexander's army, as well as young Iranians organized into a special 
bodyguard on a Macedonian model (Arrian VII, 6, 1). Berve has discussed the mutual 
adoption of practices by Macedonians and Iranians in some detail to which little need 
be added. 9 

To conclude that Alexander was deliberately seeking to counter the power and 
influence of his Macedonian officers in favor of Iranians is hardly correct, for the 
conqueror was equally capable of removing those whom he came to mistrust. It is 
most likely that Alexander trusted only those who were fully beholden to him for 
their positions. The incorporation of local troops into the Macedonian army, of 
course, was a necessity, especially after the severe fighting in Central Asia and before 
the campaign in India, since losses, and the need to settle troops in garrison towns, 
called for more and more reinforcements. So the policies of Alexander were based on 
expediency and a desire for stability to be achieved by placing in military and 
administrative positions only those who were personally loyal to the conqueror, and 
who quite naturally adopted many trappings of the Achaemenid rule Alexander had 
displaced. Yet Alexander did not have to copy the past, for he must have considered 
himself greater than the Achaemenid kings, since he had not only defeated Darius but 
had conquered more territory than they had held. 

At the end of Alexander's life many of the satraps he had confirmed in office had 
been changed. 10 In Babylonia Mazaios had died and for a time Stamenes had followed 
him to be succeeded about 324 B.C. by Arkhon, a Macedonian. At that time also the 
Persian satrap of Susiane was executed because of failure to come to the aid of 
Alexander on his march back from India, and was replaced by Argaios, then by 
Koinos, a Macedonian. In Persis, Phrasaortes had been succeeded by Orxines who was 
executed by Alexander on his return from India for his mismanagement of the 
province and again a noble Macedonian, Peukestas, was appointed. In Media, Oxydates 
had been made satrap by Alexander but was replaced c. 328 B.C. by Atropates, who 
made such an impression by his length of rule that the northern part of his realm 
took his name, today Azerbaijan. 11 Parthia-Hyrcania remained in the hands of 
Phrataphernes throughout the life of Alexander and only after the death of the former 

H. Berve, "Die Verschmelzungspolitik Alex- the Great, the main problems (Cambridge, 1966). 
anders des Grossen," Klio, 31 (1938), 135-68. His 10 For the following, the reference work by 

statement on p. 1 47 "Neu dagegen ist, dafi an der Berve, Das Alexanderreich [n. 1 ] is used and no page 

telle lokal-einheimischer Elemente Manner aus references are given since they are easy to find. 

em persischen Adel herangezogen werden," is ' ' Z. I. Yampolskii, "Ob Atropate-sovremen- 

not apparent, since the distinction between 'Per- nike Aleksandra Makedonskogo," VDI, 2 (1974), 

Ti\ a *!f Iranian nobil "y had more or less 176-77, suggests that Atropates had not been a 

co apsed by the end of the Achaemenid Empire. satrap of Media under Darius III but an indepen- 

F%k L hC StUdies by F Ham P'. W. W. Tarn and dent ruler, which may be correct. 
scn achermeyr in G. T. Griffith, ed., Alexander 



144 Chapter VI 

(c 321 B.C.) did a non-Iranian Philippos take his post, and apparently the Caspian areas 
of the Amard and Tapur people were joined to Parthia-Hyrcania. About 328 b.c. 
Stasanor was satrap of Aria or the Herat area plus Seistan, while Sibyrtios, former 
governor of Kerman, was given the combined satrapy of Gedrosia and Arachosia 
while Tlepolemus took over Kerman. The satrapy of the Parapomisos or Hindukush 
had been under Greco-Macedonian control as had been India. Thus we find a drastic 
change from the beginning of Alexander's conquests in the east, but we do not know 
why. Was Alexander following the advice of his Macedonian officers? This is 
unlikely since he continued until the end of his life to favor Iranians, including their 
clothing and customs. Did the change mean that he could no longer find Iranians 
whom he could trust in high positions, or did he realize that a division of authority in 
the provinces between Greco-Macedonian military and Iranian secular leaders would 
not work? The order, early in 324, to the satraps and commanders to disband all their 
mercenary forces indicates that Alexander was fearful of possible revolts against his 
authority. In any case, the result of Alexander's actions was to pave the way for the 
retention of power by the Macedonians after his death such that the Hellenistic age 
brooked no competition from Iranian dynasts trying to exert their influences on the 
main course of history. Likewise the release of many mercenaries only made the 
Hellenistic age one of competing generals hiring mercenaries to fight their battles. For 
the conquests of Alexander had shown the success of well-organized professional 
armies against any popular, conscript armies. The masses of different local levies could 
not stand before a small, tightly knit army, and Alexander had put the seal on that 
development. 

Some scholars have supposed that the satrapy of Persis continued to pay no taxes, as 
under the early Achaemenids, but there is no evidence that this privilege was either in 
force at the end of the Achaemenid empire, or existent under Alexander. The satrap 
of Persis, Peukestas, was infamous among Macedonians for his adoption of Persian 
customs and learning the Persian language (Arrian VI, 30, 3 and VII, 22, 3). 

THE AFTERMATH OF ALEXANDER'S DEATH 

The great conqueror died in early June 323 at the age of thirty-two at Babylon, and 
the question of continuity was pertinent. The administration of the empire had 
continued the Achaemenid model with satrap, military commander, and tax 
collector as the three important divisions of provincial authority, and at the time of 
Alexander's death from the sources we infer that non-Iranians had replaced Iranians in 
these offices in most of the provinces. Whether this represented a new policy change is 
uncertain, but most people were reconciled to this state of affairs, and Iranians had 
been integrated into the army as well as into the bureaucracy. There was neither any 
pretender to the Achaemenid throne nor any great center of resistance to the orct- 
which Alexander had imposed. For the Classical scholar inner-Greek and Macedonian 
conflicts are of prime importance, but for Iranians both were their rulers, and in the 
context of Iranian history they are combined as the Greco-Macedonian conquerors. 
Some scholars have suggested that Alexander's close friend Hephaestion was a prime 
minister or vizier of the empire on the basis of Arrian's description of him as 
X^iapxos im rrj lttttu) rrj eraipiKfj (VII, 14, 10), but the title 'commander of the 



Alexander the Great and the Seleucids 145 

Companion cavalry' as well as what we know of Hephaestion's activities do not show 
him exercising wide administrative functions which one would expect in a vizier. 12 
On the other hand, we do have the express statement of Diodorus (XVIII, 48, 4) that 
the title chiliarch (Old Iranian *hazarapati-) meant second in authority in the 
Achaemenid Empire, but it seems more likely that the military position of the 
chiliarch or 'chief of the royal guard' became second in importance after the king at 
the end of the empire, rather than to suppose that an administrative position 
equivalent to the vizier of Islamic times always existed under the Achaemenids. It has 
been argued in great detail that Xerxes, when he said of his father in his Persepolis 
inscription (XPf, line 31) pasa tanitm mam madikam akunauVaftei himself made me 
the greatest,' was in fact referring to the position known from later times as the 
'second after the king.' 13 Since Xerxes was the crown prince, and the inscription does 
not imply an existing title, it would be stretching the information provided by the 
statement of the inscription into an established position equivalent to the later vizier. 
It is more likely that Diodorus is referring to the great importance of the chiliarch 
under the later Achaemenids rather than to an established administrative position. 
Such, it seems, was the case of Hephaestion, who was Alexander's alter ego, but not his 
vizier. 14 The personal seal of Alexander, on his deathbed was given to Perdikkas 
(Arrian XVII, 117,4; Curtius X, 5, 4) which, however, did not indicate that the close 
relationship which Hephaestion had with Alexander was transferred to Perdikkas. It 
also did not mean that the latter succeeded Alexander, as we well know from 
following events. Under the eyes of Eumenes of Kardia, the secretary of King Philip 
and then of Alexander, we may assume that the Achaemenid bureaucracy continued, 
with, however, two new features, the introduction of the Greek language and the 
military colonies and towns founded by Alexander. 

The process of 'Hellenization' many times has been compared with the 
'Westernization' of the Orient in modern times. How much one should push this 
analogy is open to question, but the position of the Greek language in the successor 
states of Alexander's empire does bear a resemblance to English in the latter part of the 
twentieth century. Just as today knowledge of English has been adopted by a new 
'middle' class of Iranians, so at that time Greek served a similar function, and language, 
rather than religion, race or any other criterion, was the key to distinctions. Greek had 
started already in Achaemenid times as a kind of lingua franca in the western part of 
the empire, and Alexander's conquests spread the use of the language to India and 
Central Asia. To be sure, as a written language the use of Aramaic continued on the 
Iranian plateau, but clearly Greek was more widespread as the imperial language of 
Alexander and his successors. 



,2 Cf. E. Badian, "The Administration of the 14 I follow Berve,D<t$ Alexanderreich [n. 1], 112, 

Empire," Greece and Rome, 12 (Oxford, 1965), 176. 320, when he says that the 'chiliarch,' especially 

1 3 E. Benveniste, op. cit. [ch. 5, n.74], 64-65. The Hephaestion under Alexander, "scheint durchaus 

assertion of J. Junge, "Hazarapatis," Klio, 33 als ein Hofamt, nicht als Verwaltungsorgan 

(1940), 29, that the chiliarch under the Achaemen- gedacht gewesen zu sein." This did not mean that 

ids was the chief of the treasury and of records is the influence or authority of the 'chiliarch' was 

not supported by Elamite documents from diminished, rather the opposite since it was not a 

Persepolis. limited or fixed office. 



146 Chapter VI 

The founding of garrisons and cities (poleis) has been much discussed. 15 Certainly 
military garrisons were established by Alexander in many parts of the Iranian plateau 
but whether he also founded any Greek cities (polis) is debated, for the evidence seems 
to be only in their names - all Alexandria. Many scholars have argued that any city in 
the east with the name Alexandria must have been a full-fledged polis, but several 
questions arise; did Alexander found each one or has his name subsequently been 
added? Did Alexander establish a polis, or was it in each case a military garrison which 
later became a polis? When Tarn, following Tscherikower, asserts that Alexandria in 
Aria, which he equates with Herat, without doubt was founded by Alexander 
because Strabo and Pliny both say so, this is hardly decisive proof of its founding by 
Alexander. 16 The historians of Alexander do not mention the founding of a city 
there, but they do tell us that Alexander had to suppress a revolt of the people of the 
area under Satibarzanes. Perhaps this revolt induced Alexander to establish a garrison 
which later developed into a city. Why Tarn says that neither Seleucus nor "any 
Successor ever used the Alexander-name" escapes me, since in Appian's Roman 
History {Syriake, 57) the author specifically says that Seleucus founded two cities in 
honor of Alexander. 17 According to Arrian (e.g., VI, 15, 2; 15, 4) Alexander did 
build or order to be built certain cities; why he does not mention Herat and 
Prophthasia in Seistan is unknown, but when Plutarch (Moralia, De Alexandra 
fortuna, 328F) says there were five Greek cities which would not have existed save for 
Alexander's conquests, this does not mean that Alexander founded them, as Tarn 
supposes, for Seleucia on the Tigris, one of the cities mentioned, we know was not 
established by the conqueror. 18 It is not here proposed that Alexander founded no 
cities in the east, but rather that the Alexander historians must give us clues, rather 
than later authors who give information from various sources each of which must be 
checked, and the Alexander legend was already in vogue. 

The city called Alexandria ad Caucasum is attested by all of the historians and may 
be accepted as an authentic foundation of Alexander, who settled both natives and his 
soldiers in it (Arrian III, 28, 4; Curtius VII, 3, 23). Tarn further declares that an 
Alexandria in Arachosia, mentioned by later authors (Isidore of Charax, Ptolemy, 
etc.) must be present Ghazna, but this is questionable in view of recent excavations at 
old Qandahar where an Achaemenid settlement and a Hellenistic town have been 
partially excavated. 19 From the lack of early coins found at Begram by French 
excavators, it would seem that this site, which was identified as Kapisa, was not the 
same as Alexander's foundation, but it must have been nearby, since the valley to the 
north of Kabul, today called Koh Daman, 'the skirt of the mountain,' though not 



15 V. Tscherikower, "Die hellenistischen Sta'dte- implications about the city although possible are 
grUndungen von Alexander dem GroBen bis auf by no means proven as Tarn implies. 

die Romerzeit," Philologus, Supplementband XIX, ,8 Tarn, Creeks (ch. 1, n. 23], 14, 347, 482. 

Heft 1 (Leipzig, 1927), esp. 96-106, and W. W. 19 Ibid., 470; D. Whitehouse, "Excavations at 

Tarn, "Alexander's Foundations," in Alexander the Kandahar, 1974," Afghan Studies, 1 (London, 

Great, 2 (Cambridge, 1 948), 232-49, and the first 1 978), 9-11, 33-34 ; also A. McNicoll in ibid., 41- 

chapter of his Greeks (ch. 1, n. 23]. 44. In Ptolemy VI, 18, 4, the town of Gazaka or 

16 Tarn, Alexander, 2 [n. 6], 234 ; Tscherikower, Gauzaka (Ganzaka) is likely to be the site of later 
op. cit. [n. 15], 102. Ghazna. 

"Tarn, op. cit. [n. 6], 238. His further 



Alexander the Great and the Seleucids 147 

extensive is large enough to hold several towns. 20 Other cities were founded in the 
east but attempts to identify them have been erudite but not convincing. 21 Another 
city is Alexandria in Bactria identified as modern Termez, while Alexandria on the 
Tanais (Eskhate = on the Jaxartes) is attested in Arrian, but the certain identification 
of this city as Khodjent (Leninabad) is not supported by excavations which have not 
revealed strata as early as Alexander. 22 Nonetheless, the foundation of certain cities 
and garrisons by Alexander can be taken as highly probable, and the aim of most of 
them obviously was to insure control of strategic routes and commerce, and to hold 
together the empire. 

The question arises whether each of the cities founded by Alexander in the east was 
really a Greek polis with all that this implied, and therefore that the conqueror was 
consciously striving to bring "Hellenism" to the east by establishing Greek cities 
throughout his conquests. This point of view has been contested on the grounds that 
Alexander did not seek to establish any exclusively Greek polis, but rather only towns 
or military colonies where Greeks, Macedonians and natives would be settled. 23 The 
realization of the garrison nature of their settlements may have come to Greek 
colonists only when Alexander had founded such settlements in the east and had 
started on his way from India back to Babylon, and this may have been a prime reason 
for the revolt of the Greek colonists in Bactria in 325 and in 323 B.C., recorded in 
Diodorus (XVII, 99, 5-6; XVIII, 4, 8; 7, 1-5) and Curtius (IX, 7). From the sources, it 
is difficult to determine whether there was one uprising before Alexander's death or a 
series extending into the post- Alexander era; in any case, some Greek colonists 
revolted and wanted to return home, but they were put down by Macedonian troops. 
Whether Koshelenko (pp. 71-74) is correct in attributing their discontent to a failure 
of Alexander to create Greek cities (poleis), where Greek citizens would completely 
dominate the government, thus causing them to revolt, cannot be determined, but it 
could have been a factor in the revolt. It would seem then that Alexander himself did 
not establish any polis in the east but later rulers did, and this will be discussed below. 
There is no evidence that Alexander intended to found independent towns which 
were not under the local satraps, and there is no reason to attribute this policy to him. 
The revolt of the Greek colonists, however, does show that many settlements in the 
east at least were made by Alexander to safeguard the empire. The analogy with 
military garrisons established by the Arabs in their conquests so many centuries later 
is striking, and one may assume parallel concerns about the safety of the empire and 
the caliphate. 

Another matter of concern to Iranians was the fiscal policy of Alexander, and it is 
clear that at some time during his rule he decided to issue silver coins on the Attic 

See J. Hackin, Nouvelles recherches archiologi- " G. A. Koshelenko, "Vosstanie Grekov v 

ItiestBegram.MDAFA, 11 (Paris, 1954), 309, and Baktrii i Sogdiane," VDI, 1 (1972), 59-78. The 

R^ Ghirshman, Begram, MDAFA, 12 (Cairo, author analyzes the Greek theories of city founding 

,9 J 6 ). 85. The site of Saray Khwaja, called by the of Aristotle and Isocrates, and comes to the 

inhabitants Iskandariya, may possibly reflect a conclusion that Alexander had no intention of 

settlement of the conqueror. following these ideas in his settlements. An 

22 Jf rn ' °P- "'■ l ch - !• n 23 1- 525 - expansion of this article appears in his book 

See B. Gafurov, Tadzhiki (Moscow, 1972), Grecheskii Polis na ellinisticheskom Vostoke (Mos- 

, ' note 13 - The identification of Alexandria on cow, 1979), 181-221. 
the Oxus with Termez is still disputed. 



148 Chapter VI 

standard for the empire, while at the same time to allow local coinage to continue. He 
even struck gold darks and issued double darks, and he followed his father in 
maintaining a value of 10 to 1 instead of the Achaemenid 13-J to 1 ratio of silver to 
gold. 24 Alexander did not mint coins in Iran or Central Asia, because there had been 
no mints there under the Achaemenids, and Babylon and perhaps Susa are the 
easternmost mint places. It is dubious whether Alexander himself had coins struck 
with his own portrait with the legend 'Alexander king,' and even at Susa, which may 
have been an Achaemenid mint town, the first such coins seem to be post-Alexander 
in date. 2S The introduction of an imperial coinage to the east must have had great 
influence in areas where formerly barter and bullion had prevailed, but we can only 
infer this from coin hoards and the profusion of coinage in the east after Alexander. 

There is no evidence that Alexander the Great had the time to initiate great changes 
in the east, and we may assume that his successors were those who laid the foundations 
of Hellenism in the East. Undoubtedly Alexander did initiate military settlements 
which later developed into cities, but his task was to hold the empire he had formed 
and at the least he seems to have conciliated most of the Iranian aristocracy who 
accepted him as their ruler. His Macedonian satraps, especially Peukestas of Persis, on 
the whole were accepted and obeyed by the local population, and for the most part 
Macedonian rule proceeded without significant revolts against it. 26 The sources tell 
us of political events and about the quarrels between the Macedonian generals, but 
not about the process of Hellenization in the east which must be reconstructed from 
rare, scattered notices in the sources and from archaeology. The wars of the Diadochi, 
as the successors of Alexander are called, had no great effect on the mass of the 
population of the Iranian plateau, but a survey of the succession insofar as it did 
influence Iran should be made. 

The first assignment of satrapies was made by Perdikkas in 323 B.C. after the death 
of Alexander, and he left all the satraps of the east intact but appointed Peithon son of 
Krateuas, a noble Macedonian, satrap of Media while Atropates continued to rule in 
Azerbaijan. One of the first tasks of the new satrap was to command an army sent to 
suppress a revolt of Greeks settled in Bactria which he did according to Diodorus 
(XVIII, 7, 5). Whether this revolt was a continuation of the previous one or not we 
cannot determine, but according to the sources, the Greeks had wished to return home 
against the wishes of Perdikkas and were suppressed. Apparently a Macedonian called 
Philip became satrap of Bactria and Sogdiana at this time. Oxyartes, father-in-law of 
Alexander had been made satrap of the Hindukush region by Alexander, succeeding 
several short-lived satraps, and he continued in his post as did several vassal kings in 
India. Eumenes was made satrap of Cappadocia and Paphlagonia in Asia Minor where 
he had to fight against native dynasts. Plutarch (Life of Eumenes, 4) adds that Perdikkas 
sent Eumenes against a certain Neoptolemos, satrap of Armenia whom he defeated, 
but no more information exists about Armenia at that time. 

24 Bellinger, op. cil. [n. 4], 30-31 , mints 1 14-27; 26 There is a curious report that Stasanor, satrap 

G. K.Jenkins, Ancient Greek Coins (N.Y., 1972), of Herat and Seistan, attacked the customs of the 
211-16; G. Kleiner, "Alexanders ReichsmUnzen," local inhabitants who then revolted against him 
Abh. der deutschen Akad. der Wiss. zu Berlin, Jhg. (Porphyrios, De Abslinentia, IV, 21). 
1947 (Berlin, 1949), 9-23. 

5 G. Le Rider, Suse sous les Sileucides el les 
Panhes, MMAI, 38 (Paris, 1965), 45. 



Alexander the Great and the Seleucids 149 

Shortly after the assassination of Perdikkas in 321 B.C., at a conference at 
Triparadeisos in north Syria, Antipater assumed the role of regent or representative of 
Philip Arrhidaeus, half-brother of Alexander, and Alexander's posthumous son, both 
of whom had been recognized as joint rulers of the empire even though few of the 
generals paid any heed to one or the other. Few changes in eastern satrapies were made 
by Antipater except that Philip was transferred from Bactria, and he replaced the 
perhaps dead Phrataphernes in Parthia and Hyrcania, while Stasanor of Soli on 
Cyprus, former satrap of Herat and Seistan, became satrap of Bactria and Sogdiana, 
while another Cypriote called Stasander became satrap of Herat and Seistan (Diod. 
XVIII, 39, 6). The capable general in charge of Antipater's army was Antigonos, the 
'one-eyed.' The entire situation changed, however, with the death of Antipater in 319 
when he bequeathed his office to Polyperchon, who was friendly to Eumenes and 
who had been attacked by Antigonos in Asia Minor. This turn of events enabled 
Peithon, satrap of Media, to attack Parthia and replace Philip with his brother 
Eudamos (Diod. XIX, 13, 7). 27 Peithon was defeated in turn by a coalition of eastern 
satraps, and then Peithon went to Babylon which had been allotted to Seleucus at 
Triparadeisos. Seleucus, however, did not join Peithon, who was now faced with the 
coalition which followed Eumenes (and Polyperchon in Europe) and wintered in 
Susiane in 317 B.C. Since the members of the coalition quarrelled about the leadership, 
Eumenes advised that all of them should participate in military decisions, which 
proved to be his undoing. 

Meanwhile Antigonos was gathering forces in Mesopotamia to attack the 
coalition, and he hoped to do this with the support of Peithon and Seleucus. He 
advanced to Susa, and Seleucus on behalf of Antigonos besieged the citadel. From the 
sources we learn that Iranians fought on both sides, one of Antigonos and the other of 
Eumenes, which latter might be called the 'royalist' camp, since they ostensibly 
upheld the unity of the empire under the son of Alexander, for his half-brother Philip 
Arrhidaeus had been executed in 317 at the order of Olympias, mother of the dead 
conqueror. At first Eumenes seemed to have the upper hand; at a battle on the Iranian 
plateau near Gabai (Isfahan) against Antigonos he prevailed, but Antigonos held a 
unified command while the coalition was disunited (Diod. XIX, 26-7), and in a 
second battle Antigonos captured the baggage of Eumenes together with the wives 
and children of many of his troops. The satraps thereupon wished to retire to their 
satrapies and did so, while the Macedonian mercenaries seized Eumenes and turned 
him over to Antigonos in return for their baggage and families (Diod. XIX, 42). 
Whether the account of Diodorus is accurate or not cannot be determined, but in any 
case Eumenes was executed and Antigonos now became lord of Asia. 28 

In the account of the forces of Eumenes, Diodorus (XIX, 27) has an interesting 

The text says he put to death Philotas, the troops and elephants from India. Oxyartes, satrap 

general (<TTpaT7jyos) of Parthia and replaced him of the Hindukush region, sent a general Androba- 

with his brother, but we have seen that Philip was zos with troops, and Peukestas of Persis was 

t e satrap. Whatever the confusion in names, recognized as the leader of the coalition. 

Peithon secured Parthia to his support. The satraps 28 On the title 'king of Asia,' probably assumed 

who sent troops for the coalition against Peithon by Antigonos after removing Peithon, satrap of 

wereTlepolemosofKerman.SibyrtiosofAracho- Media, see H. W. Ritter, Diadem uni Kdnigs- 

su. Stasander of Herat and Seistan, with extra herrschtft, Vestigia (MUnchen, 1965), 102. 
troops from Bactria, while Eudamos brought 



150 Chapter VI 

remark, that five hundred soldiers from the Hindukush area joined the coalition of 
Eumenes against Antigonos and an "equal number of Thracian colonists from the 
upper colonies (satrapies)" laoi &p&K€s ix ru>v ava> KaroiKiwv. The indication that 
Thracians were also settled in the eastern 'upper' satrapies (medieval Khurasan), as 
well as Greeks and Macedonians, is a welcome item, helping to explain the large 
numbers of colonists who came from Europe and Asia Minor to the east. One 
wonders whether special inducements of payments in money or kind, as well as land, 
were not offered to settlers to come to the east to supplement the veterans and forced 
settlement of military cadres in garrisons. For the numbers of colonists sent to the east 
must have severely depleted the populations of both the towns and countrysides in the 
west. The Hellenistic expansion to the east at this time may be compared to the earlier 
Greek settlements in the Mediterranean and Black Sea areas. 29 

Antigonos, after the death of Eumenes, began to consolidate his position in the east, 
and his erstwhile ally Peithon was suspected of rebellion and executed, whereupon his 
other ally Seleucus, fearing a similar fate, fled to Ptolemy in Egypt in 316 B.C. An 
Iranian, Orontobates, was appointed satrap of Media by Antigonos in place of Peithon. 
The satraps of the farther east were in too strong positions to be removed, so 
Antigonos retained Stasanor in Bactria, but Stasander in Herat and Seistan was 
replaced by Evitus, who died shortly thereafter and was followed by Evagoras. 
Sibyrtios retained Arachosia, and according to Diodorus (XIX, 48, 4) Antigonos sent 
to him many of the Macedonian veterans who had turned over Eumenes to 
Antigonos, with orders to send them on difficult missions where they would be 
killed, since he did not trust them. Peukestas, the popular satrap of Persis was removed 
and Asklepidoros put in his place, and another Peithon, former satrap of India, was 
now made satrap of Babylonia, while in Susiane a certain Aspeisas was installed. 30 
Thus, Antigonos secured control of the east, but he remained suspicious of the fickle 
satraps who would only support him if it were to their respective advantages. Then 
Antigonos turned to the west, but a defeat of his son Demetrius by Ptolemy at Gaza, 
Palestine, in 31 2 persuaded Seleucus to try his luck again in the east, and Ptolemy gave 
him a small force of soldiers to assist him. Seleucus had been well liked by the 
Babylonians when he previously had been their satrap, so many welcomed him back, 
and in honor of his entry into Babylon in the late summer of 312, although according 
to Babylonian reckoning the first of the month Nisan or the 3rd of April 31 1 B.C., this 
date was the beginning of the Seleucid era. 31 Dating in this era in Babylonian clay 
tablets is attested, however, only from the year 304/3 B.C. 

Seleucus devoted himself to consolidating his position in the east, but Nikanor, the 



!, The problem of the colonists will be i\ htruzione Chssica nuova serie 10, (1932), 

discussed in more detail below. From the sources it 462-66. 

is clear that Iranians fought on both sides. Cf. infra, 3I J Oelsner, "Keilschriftliche Beitra'ge zur 

n - '5. politischen Geschichte Babyloniens in den ersten 

The coin with the name Aspeisos may well Jahrzehnten der griechischen Herrschaft (331-305 

refer to this satrap, but it is unusual; cf. Bellinger, v.u.Z.)," AF, 1 (1974), 135. The Seleucid era was 

op. cil. [n. 4], 88-89. On the struggle between commonly called the 'era of Alexander .'Cf. Chroti. 

Antigonos and Seleucus, see G. Furlani, "La cronaca Maroniticum, CSCO, Script. Syr. ser. Ill, IV, pt. 1 , 



babilonese sui diadochi," Rivista di Filologia e 43. 



Alexander the Great and the Seleucids 151 

eneral in charge of Antigonos' forces in Iran prepared to attack Seleucus, while 
Demetrius, son of Antigonos, invaded Babylonia and recovered some of the places 
Seleucus had taken, but he could not assure control of the satrapy for Antigonos, since 
he had to return to the west and to the affairs of mainland Greece. In 310 the widow 
of Alexander the Great, Roxane, and her son Alexander were murdered by Kassander, 
and any possible succession to the empire of the conqueror was ended. Demetrius, son 
of Antigonos, was only partially successful in Greece, but in 306 he destroyed the 
navy of Ptolemy near Cyprus and captured the island, and after that both Antigonos 
and his son assumed the title of 'king' and both wore diadems. In 304 B.C. Ptolemy, 
Seleucus, Kassander and other Diadochi also took the title 'king,' and any pretence of a 
united empire was now ended, and the heritage of Alexander was permanently 
divided. 

Meanwhile Seleucus had in effect replaced Antigonos as ruler of the east, especially 
after defeating Nikanor in a battle in which other followers of Antigonos lost their 
lives (Diod., XIX, 92, 5). Appian (Roman History, Syriake, 55) says Seleucus killed 
Nikanor in battle, but he does not say when or where. Seleucus must have either 
conciliated the satraps in the east or defeated them, but the only item of information 
we have is that Seleucus made peace with the new Indian power headed by 
Chandragupta (SavSpoKorros) and obtained from him five hundred elephants in 
exchange for recognition of Mauryan supremacy at least over the Indus basin if not 
more (Strabo XV, 724). Before this time, according to Will, Seleucus had defeated the 
forces of Antigonos in a great battle, probably in 309/8 B.C., which resulted in a treaty 
by which Antigonos renounced rule over Iran. 32 This, however, may be the battle 
against Nikanor in Babylonia, but the existence of a treaty between Antigonos and 
Seleucus is an assumption of Will based on the absence of records of further conflict 
between the two. In any case, we are left to conjecture just how Seleucus won the east 
and what were his relations with Antigonos. 

The end of Antigonos came in the summer of 301 B.C. at the battle of Ipsos in north 
Syria, when a coalition of forces under Kassander from Greece, Lysimachos and 
Seleucus, with his newly acquired Indian elephants, crushed Antigonos, who died on 
the field of battle, while Demetrius, his son, fled with other forces. 33 The three victors 
divided the domains of Antigonos between them, and Seleucus received all of the east 
and Syria. Southern Syria or Palestine had been occupied by Ptolemy, who had not 
taken part in the battle of Ipsos, but Seleucus for the time acquiesced in the occupation, 
although he did not renounce his claims to the south. Lysimachos took most of Asia 
Minor. The intrigues and problems of the west do not concern us, but inasmuch as it 
took the attention of Seleucus away from the east, in this manner Iran was influenced 
by events in the west. What do we know about Iran and Central Asia at the beginning 
of Seleucid rule? 



Will, supra, Histoire politique, 58. This view is 33 O. Miiller, Antigonos Moitophthahnos und "das 

contested by C. Wehrli, Antigone et Demetrios Jahr der Konige," Saarbrlicker Beitrage zur Alter- 

(Gerieva, 1969), 130, who denies that there was a tumskunde. 1 1 (Bonn, 1973), 140 pp., discusses 

«tle or treaty, but merely that Antigonos questions of legitimacy and kingship in the career 

relinquished rule in the east to concentrate on the of Antigonos. 
west. 



152 Chapter VI 

THE CONSOLIDATION OF THE SELEUCID KINGDOM 

If we accept a statement of Diodorus (XX, 53, 4) that Seleucus took the title of 'king' 
shortly after he had taken over the 'upper provinces,' the date of this conquest would 
be c. 307-306 B.C. or later. As mentioned, we hear nothing of Seleucus in the east save 
his ceding of territory to the founder of the Maurya dynasty Chandragupta, but the 
extent of that ceded land is uncertain. A chance remark of Justin (XV) that Seleucus 
subdued the Bactrians may indicate some fighting before obtaining the allegiance of 
that province. This is all that has survived, but we may suppose that most of the 
inhabitants of the east, natives and Greco-Macedonians, supported Seleucus, since no 
reports of revolts or loss of territory survive from the reigns of either Seleucus or his 
son Antiochus I. The wife of Seleucus and mother of Antiochus I was an Iranian, 
daughter of the Bactrian leader Spitamenes, who had fought Alexander, so we 
suppose a certain sympathy for the Iranians in the first two Seleucid rulers which 
engendered a loyalty to them for a long time. At the same time neither Seleucus nor 
Antiochus relied primarily on the native population of their vast domains to support 
their rule. Just as the other Diadochi, so the Seleucids also relied on Greco- 
Macedonian colonists as their basic support, and they carried on the work of city 
founding started by Alexander and actively pursued by Antigonos. 34 It seems 
apparent that both Seleucus and Antiochus exerted themselves greatly not just to 
plant garrisons but to make Hellenistic centers on the routes to the east so they might 
be held by settlers loyal to their new monarchy. 

Antioch on the Orontes was founded by Seleucus in 300 B.C., a year after the defeat 
of Antigonos, and it became first the overall then the western capital. A few years 
later the eastern capital Seleucia on the Tigris was founded and it was settled primarily 
with people from Babylon. 35 Apparently this made the Babylonians unhappy, but 
the decline of their city was assured, when Antiochus I in the thirty-seventh year of 
the Seleucid era (275 B.C.) made Seleucia the 'royal city,' which then made of Babylon 
only a place of pilgrimage. 36 The ancient city of Raga or Rhages was re-founded by 
Seleucus, or much more likely it had a large influx of Hellenic settlers, and thereafter it 
was called Europos, according to Strabo (XI, 524; also Ptolemy VI, 2, 17). The later 
capital of the Parthians Hekatompylos (Komis/Qumis), according to Appian (Syr., 
57) was also founded by Seleucus, but no other evidence for this or other cities in 
Parthia - Soteiria, Kalliope and Kharis - all supposedly founded by Seleucus exists, 
although Stephan of Byzantium says the first was founded by Antiochus I. These 
cities, named with Greek epithets are impossible to localize precisely without the 
native names. In Bactria and Central Asia the first two Seleucids undoubtedly 
established colonies but we cannot determine where. Tscherikower's conclusion that 
whereas Alexander and Antigonos primarily had military reasons for their 



4 For the settlements of Antigonos see C founding is noted by Appian. Roman Hist., Syr., 

Wehrli. op (it |n. 32|. 1 33, but in the east we find 57 

few cities founded by hint unless their names were l " lor references to clay tablets see S A. I'allis. 

changed by later rulers "The History of Babylon 538-93 B(..." in Stadia 

On Seleucia, sources, etc , see Tscherikower, Orwnltilia loani Pi'dersen, ed by F. Hvidberg 

op rit |n 15|, 90 The prodigious activity of city (Copenhagen, 1953), 282 



Alexander the Great and the Seleucids 153 

foundations, but the Seleucids were more interested in trade and commerce, is very 
attractive and convincing. 37 Antioch on the Orontes and Seleucia on the Tigris were 
the two most important foundations of Seleucus, controlling the 'Fertile Crescent,' 
but other cities were completed along a line of communications to the east. 

The process of settlement in the east must be inferred from information on city 
foundations in Syria and the west, since there is no information in the sources about 
eastern cities. The excavation of a Hellenistic city on the banks of the Oxus River in 
Afghanistan called today Ay Khanum, however, has provided most welcome 
information to compare with cities in the west. Ay Khanum at first was tentatively 
identified with the city of Eucratideia, named after the Greco-Bactrian ruler 
Eucratides (Strabo XI, 576 ; Ptolemy VI, 11,8) which would date its founding in the 
third century, but it seems that it had an earlier foundation perhaps from the time of 
Seleucus I. 38 In any case, there was a Hellenistic city on the upper Oxus River where 
the Kokcha joins it, and the remarkable Greek inscriptions, the theatre, gymnasium, 
agora and city plan which have been recovered by the archaeologists, all emphasize 
the thoroughly Greek character of the settlement. One inscription, among others, 
implies that a certain Clearchus, who wrote books as well as epigrams, one of which is 
on the base of a stele from the site, travelled from Delphi in Greece to this outpost in 
the east. 39 Hermes and Herakles are mentioned in another inscription, further 
evidence of the profoundly Greek character of the city. Apparently later, perhaps c. 
1 50 B.C. in the Greco-Bactrian period, a (royal) court was added in the city, unusual 
for Hellenistic cities. The discovery of small fragments of philosophical papyri in 
Greek also attest to the strong Greek intellectual presence in a city far from Greece, 
but above all it indicates that the settlers were not just military colonists, but cross 
sections of all of the population were represented in the Hellenistic colonization of the 
east. 40 

Seleucus followed Alexander in many respects, one of which was the policy of 
sending expeditions to far away provinces to learn about the limits of his domains. 
Patrokles was a Greek high official in the court of Seleucus who explored the coasts of 
the Caspian Sea, perhaps seeking water routes to the east since he also sailed in Indian 
waters (Strabo II, 69, 74; and Pliny VI, 58). A general of Seleucus, called Demodamas 
ot Miletus, may have led an expedition across the Jaxartes River at the end of Seleucus' 
reign (Pliny VI, 49) and Megasthenes, followed by Daimakhos (or Demimakhos) 
maintained an embassy at the Mauryan court of India on behalf of Seleucus, and it 
would seem that in spite of the ceding of territory to the Indians by Seleucus friendly 



■ ' Tschenkower, op cit [n. 15]. 169. 1977 a AT Khartoum," CRAl (avril-juin, 1978), 

L. Robert apud V Bernard. I'ouillcs d'Ai 458, adds that Clearchus of Soloi, who visited Ay 

Minimum. MDAFA. 21, 1 (1973), 217-22. calls Khanum about 300 B.c and was a follower of 

Kineas, probably from Thessaly, the founder of the Aristotle had two elegaic couplets of precepts at 

city, since a Greek inscription has been found there. Delphi erected in the temenos of Kineas in the city, 

presumably over the tomb of Kineas, in the agora •*" Ibid. 459-60 On funerary jars the names 

° ,A V Khanum, and Robert suggests the town was Lssamos and lsidora (worshipper of Isis?) were 

•'" Alexandria founded by the conqueror, al- found, and the dedication of the gymnasium was 

though previously (p 210) he had attributed the nude by two brothers Triballos and Strato. 

toumlmg to Seleucus 1 possible of Thracian origin. 
1' Bernard, "Campagnc de fouilles 1976- 



154 Chapter VI 

relations were maintained between the two empires. 41 The amount of land ceded is 
unknown, but since Megasthenes had served in the satrapal court of Sibyrtios in 
Arachosia before being sent on the mission we may suppose that Arachosia probably 
was not included in the territory ceded by Seleucus to Chandragupta, although this 
province may have been conquered by Asoka, grandson of Chandragupta, since his 
inscriptions are found there. 42 These inscriptions, in Aramaic and in Greek, are 
fragments of copies of several of the edicts of Asoka, who was converted to 
Buddhism, and they give us an interesting picture of the population of those areas, 
once part of Alexander's empire and then under the Mauryas. 43 There were three 
groups in Arachosia, if we follow the inscriptions, Greco-Macedonians, Iranians who 
used Aramaic as their written language, and Indians. The Iranians may well be those 
people called Kambojas associated with the Greeks in the thirteenth rock edict of 
Asoka. Their location and the meaning of the word Kamboja are much debated, but it 
is at least agreed that they were Iranians living to the northwest of the sub- 
continent. 44 In the eyes of Indians, the Greeks (Yonas or Yavanas from Ionia) and the 
Iranians were associated, and this is a good characterization of the eastern Seleucid 
domain, a dual control. 

In addition to gathering information about his domains, Seleucus realized that it 
would be impossible to govern the east from Antioch on the Orontes, so he followed 
Antigonos in appointing a co-ruler of the east, in this case his son Antiochus, c. 294 or 
293 B.C. 45 Seleucia on the Tigris became the eastern capital, the seat of Antiochus. 
According to Bengtson, the leading authority on the organization of the Seleucid 
state, there were three categories of subjects of this state, the local dynasts, the cities and 
the ethne, or 'peoples,' a division inherited from the Achaemenids. 46 The ethnos should 

41 On Megasthenes the works of J W McCrm- har, see F. R. Allchin and N. Hammond, The 
die, Ancient India as Described by Megasthenes and Archaeology of Afghanistan (London, 1978), 192- 
Arrian (London, 1877), and Ancient India as 98, with bibliography. 

Described in Classical Literature (Westminster, 44 V. S. Agrawala, India as Known to Panini 

1901) are still useful as reference works, but there (Lucknow, 1953), 48, identifies them asGalchasof 

are many books on the relations such as O. Stein, the Pamirs, which is unlikely. E. A. Grantovskn, 

Megasthenes und Kautilya (Vienna, 1 922) ; cf. Will, "Plemennoe obedinenie Parca-Parcava," in Istoriya 

supra, Histoire politique, 237-38 for sources and i Kultura Drevnei Indii, ed. by W. Ruben and V. 

discussion. Strove (Moscow, 1 963), 72-77, concludes that the 

42 Although Appian (Rom Hist , Syr , 55) says Kambojas lived in Arachosia and in the Ghazna 
Seleucus ruled to the Indus River, there is no area The Parsu of Panini he identifies as ancestors 
indication how long he ruled in the east It is not of the Pashtuns or Afghans. 

unlikely that Chandragupta's son Bindusara ex- 4S Under Antigonos first Peithon (c. 323-316 

panded the Mauryan empire as did Asoka in his B.(.) and then Nikanor (315-312) had served as 

early years, so the suggestion above has much to stratCgos (or general-governor?) of the upper 

commend it. On a possible ma'rnage between provinces The title is equivalent to satrap, of 

Seleucus I and an Indian princess, see J. Seibert, course, with more emphasis on the military side, 

Historische Beitrage zu den dynastischen Verbindun- but whether any real control was exercised by him 

gen in hellenistischer Zeit (Wiesbaden, 1967), 3. over the satraps of the farthest east is dubious. 

4-1 For a description of the various inscriptions. 46 H Bengtson, Die Strategie in der hellenisti- 

I'ul-e Darunta in an Indian Prakrit language schen Zeit, 2 (MUnchen, 1944), 3, and his Grie- 

(Aramaic alphabet) and in the Aramaic language, chische Geschichte [ch 5, n. 1 8|, 438, supplanting E. 

two Aramaic inscriptions from the Laghinan Bikerman, I.es Institutions des Seleucides (Paris, 

valley, a bilingual Greek and Aramaic inscription 1 938) and M. Rostovtzelf, The Social and liconoinic 

near Qandahar, an Indian Prakrit (Aramaic History of the Hellenistic World. 1 (Oxford, 1941), 

alphabet) and Aramaic language bilingual from 423-542 
Qandahar, and a Greek inscription from Qanda- 



Alexander the Great and the Seleucids 155 

. the Old Persian dahyu- 'land' mentioned frequently in the Old Persian inscriptions. 
The Achaemenid city, OP vardana, did not have anything like the importance of the 
oolis of Hellenistic times, and it is hazardous to assert that cities on the Iranian plateau 
before Alexander had the same status as Mesopotamian cities, while the latter also 
must not be confused with the megalopolis Babylon, or with other cities of a 
millennium earlier. The last category, the dynasts, were at first local rulers who had 
been absorbed into the empire without fundamentally losing their positions. Later, in 
the Achaemenid Empire and under the Seleucids, friends of the ruler, generals of the 
army and others received land with various privileges from the ruler, and some 
became the equal of local dynasts in wealth and authority. Land acquired by temples, 
through gift or purchase grew in size and independence under the Seleucids, but the 
degree of control which local governors, or the central court, exercised over temples 
undoubtedly varied considerably in different parts of the Seleucid domain. Likewise 
the control over cities by the central government varied considerably, even though de 
jure a polis might be regarded as independent and free. In practice the ruler had the 
most to say about the government of a city, although clashes over authority with the 
provincial government also occurred in many cities. It seems that the word ethne, 
used in Egypt and elsewhere, is very rare in a Seleucid context where instead olro-noi, 
which would somewhat correspond to the Old Persian word for 'land,' appears, but 
the expression does not necessarily mean the equivalent of 'satrapy,' although it 
sometimes does. 47 

The territorial division of the Seleucid state at first followed the satrapal divisions 
of the Achaemenids and Alexander, but in time the satrapies were divided and more 
of them were created. In the east, however, Persis, Susiane and Media remained much 
as they were, while further east a progressive loss of territory occurred, first 
Arachosia, then Bactria and Parthia. Azerbaijan and Armenia, although at times 
nominally subject to the Seleucids, were really independent in practice. In regard to 
nomenclature, all scholars agree that the Greek sources are imprecise and 
contradictory, and the titles 'satrap and 'strategos' have caused dispute, but Bengtson's 
suggestion that the latter title is used only in the western part of the empire as a 
synonym for the former, and that after Antiochus III the term strategos is more often 
used in place of 'satrap' is convincing. 48 The provincial sub-divisions also cause 
trouble, for the terms hyparchy, toparchy, eparchy and meris or fiepiSapx^a- appear in 
the sources. Tarn suggested that the Seleucid satrapy was divided into eparchies, which 
in turn were divided into hyparchies, while Altheim proposed a fourfold division: 
satrapy, meris, hyparchy and toparchy, the last also called stathmoi, but Bengtson's 
proposal that there was only a threefold division of satrapy, hyparchy and toparchy, 
with a rare meris in between a satrapy and a hyparchy and headed by meridarch only in 
the far east and in Palestine, here borrowed from the Ptolemies, fits the evidence 
better than the other theories. 49 It is clear that the subdivisions were not uniform 
throughout the Seleucid domains, but we can say that the satraps in the east continued 



Bengtson, Stralegie [n. 46], 1, 10-11. Geschichte Mittelasiens im Allertum (Berlin, 1970), 

Ibid., 48-51. 

Ibid., 30-38; W. W. Tarn, "Seleucid-Parthian 

ies," Proceedings of the British Academy, 16 

(London, 1930), 24-33; F. Altheim and R. Stiehl 



* 8 Ibid., 48-51. " 325 The / term chiliarchy also appears in an 

49 Ibid., 30-38; W. W. Tarn, "Seleucid-Parthian inscription 'from Sardis, but not in the east (Tarn, 
Studies," Proceedings of the British Academy, 16 op. cit. [ch. 1, n. 23], 28). 



156 Chapter VI 

in the Achaemenid tradition of being miniature kings with courts copying the central 
court. Certainly in Bactria and Parthia the satrap wielded military as well as political 
power. In the west, on the other hand, the central army and constant warfare with the 
Ptolemies, and with the Antigonids of Macedonia and others, gave the military 
element a preponderant role in the running of affairs. Revolts in the eastern satrapies 
seem to have followed a similar pattern as in the Achaemenid Empire, but under the 
Seleucids they eventually succeeded in detaching large parts of the empire from the 
central authority. 50 

Just as the Achaemenid king and Alexander had their 'friends' at court, so did the 
Seleucids, and undoubtedly the change of rulers brought great changes in satrapal 
appointments and new individuals to the fore. Since the Seleucids were continually 
engaged in struggles in the west the generals (strategoi) at court became very 
important. In the east loyalty to the Seleucid government was based mainly on the 
K&roiKot or military colonists who were also the founders of the cities. The relation 
of these colonists to the satrapal organization is unrecorded, but possibly they had the 
same relation to the satrap as did the cities in Syria and Asia Minor to the Seleucid 
king, quasi-independent, yet usually following his orders. The various combinations 
of central and provincial troops and commanders in the west throughout the Seleucid 
period do not concern us, rather only the domains of the governor-general of the east, 
a kind of co-regency, which is not to be confused, however, with the institution of 
double kingship such as two Spartan kings or two consuls in Rome. The east, after 
Seleucus had appointed his son as co-regent, was the land east of the Euphrates and 
included Mesopotamia, but it is not known where Antiochus spent much of his time 
as co-regent, presumably on the plateau. Margiane, or the Merv oasis, was surrounded 
by a wall of 1 500 stadia (over 300 km.), and a city named Antiocheia was founded by 
Antiochus at some time (Strabo XI, 516; Pliny VI, 47). It may have been a rebuilding 
of a city founded by Alexander but destroyed by nomads, according to Pliny, but this 
was surely only one of the foundations of Seleucus or Antiochus in the east. As noted, 
it would seem that the prime motivation in the founding of cities in the east was to 
secure the line of communications and trade from the lowlands of Mesopotamia to 
Bactria and farther east. 

One study of this period emphasizes the Iranian reaction against Hellenism, which 
is contrary to the majority of writings which emphasize the spread of Hellenic culture 
in the east. 51 Eddy suggests that on the plateau Persis was the satrapy which opposed 
Hellenism the most, while Media and Bactria less so. He neglects the simple but 
striking fact that most of the Hellenistic city foundations of the Seleucids were in 
Media and on the route to Bactria, and out of the way areas were not exposed to 
Hellenistic culture to the same degree. The theory of a strong religious opposition to 
the Seleucids, based on the close connection between the Achaemenid royal house and 
Zoroastrianism, with a twin motivation, restoration of Iranian rule and religious 
supremacy, is attractive but not necessarily universally valid. For not only the 

The process of breaking away in many cases is little or no evidence, however, to support any 

may have been gradual, first a removal of central theory for the unconditional breakaway of satraps, 

officials in the province, followed by a reduction On the Achaemenid provincial divisions, see J. 

of military levies and tribute to the central court, Junge, "Satrapie und natio," Klio, 34 (1942), 49. 
then a vassal status and finally independence. There 51 Eddy, supra, King is Dead, 75-100. 



Alexander the Great and the Seleucids 157 

tolerance of the Hellenistic rulers towards local religions and cults, but the conscious 
identification of Greek with Oriental deities, the oft-described Hellenistic syncretism, 
cannot be ignored as a factor of conciliation. In the religious domain it appears that the 
westerners were more the learners and adapters, and there is little evidence of clashes 
to be found in the sources between Hellenic and Oriental religions. But again 
conditions surely varied in different parts of the Seleucid realm, which we should 
examine, beginning with Susiane. 

Susa remained as important a city under the Seleucids as it had been under the 
Achaemenids, and the first Seleucus renamed it Seleucia on the Eulaios (River) and it 
became a full-fledged Greek polis with the features of a Hellenistic city with 
magistrates, gymnasium, etc. 52 The large number of copper coins, first introduced by 
Seleucus I, found in French excavations over the years, assure us that Susa was an 
important mint for the Seleucids, and even before them, while Greek inscriptions 
from Parthian times indicate the strong persistence of Hellenism into the Christian 
era. Susiane was undergoing a Semiticization as well as a Hellenization under the 
Seleucids, and the native Elamites maintained their language and culture primarily in 
the mountains to the east, in what came to be called Elymais. The plains of Susiane 
were an extension of Mesopotamia, and the coins found at Susa from Characene or 
Mesene, at the mouth of the Tigris-Euphrates, attest the close commercial relations 
between the two areas. 53 Since the Seleucids were interested in trade and commerce 
and founded cities on the Persian Gulf, such as Ikaros on the island of Failaka near 
modern Kuwait, and Charax, founded by Alexander, we may assume that Susa 
formed part of that complex of Hellenic settlements which remained loyal to the 
Seleucids long after the Parthians replaced them on the plateau, and Hellenic 
influences remained strong afterwards. There is no need to repeat the ample survey of 
the Greek inscriptions and names from Susa, which indicate a Hellenic domination of 
the local government throughout Seleucid rule, at least until the year 164 B.C. date of 
the death of Antiochus IV. Even afterwards, until the Elymaians captured the city 
about a.d. 45, the Greek language and Greek laws and institutions flourished at 
Susa. 54 From the Greek inscriptions and the coins we also see the Greek gods 
worshipped in the city, with Artemis and Apollo predominant, perhaps in the official 
cult of the Seleucid dynasty, although other deities appear on the coins infrequently 
such as Athena, the Dioscuri, Hermes, and perhaps Zeus and Herakles (Le Rider, p. 
289). Nanaia of Susa, to be identified with Iranian Anahita, is mentioned in several 
inscriptions but the identification with Artemis seems to be a development of 
Parthian times (Le Rider, p. 295). The process of absorption of the Hellenic elements 



On the history of the city under the Seleucids 53 On the location of Charax see J. Hansman, 

** Le Rider, op. cit. [n. 25], 274-93. Le Rider ably "Charax and the Karkheh," IA, 7 (1967), 36-58. 

defends the naming of the city as "Seleucia' by For a history of Characene cf. S. A. Nodelman, "A 

Seleucus I against Tarn who favors Antiochus HI as Preliminary History of Characene," Berytus, 13 

the ruler who made it a polis. The further (1959-60), 83-121. The name Elymais comes 

suggestion (p. 281) that colonists from Ephesus from Elam. 

were settled at Susa, because the local goddess 54 Le Rider op. cit, 287. The expansion of rice 

anaia was later identified with Artemis of culture and vine culture in Susiane (the latter 

tphesus, is attractive. One should note that difficult on the hot plains) is attributed to the 

Macedonian names also occur there. Macedonians by Strabo (XV, 732). 



158 Chapter VI 

in Susa was long, and here, more than elsewhere on the plateau, except perhaps 
Bactria, the influences of Hellenism survived. 

On the plateau in the mountains of Elymais, we have no sources from the period of 
Seleucid rule to indicate either the degree of Hellenization, undoubtedly much less 
than in Susiane, or the extent of Seleucid control. Later Elymais became an 
independent state, but under the early Seleucids we can only surmise that local cult- 
shrines existed in the mountain valleys, as they did later, but central Seleucid control 
was almost non-existent. 

Persis, the favored homeland satrapy of the Achaemenids, however, presents 
problems for we do have equivocal information which is differently interpreted. As 
far as one can determine, Persis, with Persepohs, was under Macedonian rule to at least 
the early part of Seleucus' reign, mainly because of a few of his early coins found at 
Persepolis and also his coins found at Pasargadae. 55 How long loyalty to the Seleucids 
lasted is difficult to determine, for the series of local coins struck in Persis indicates an 
early rejection of Seleucid rule, which is not unexpected since Persis was the one 
province of the Achaemenid Empire which formerly did not pay taxes and where 
local opposition to the conquerors would develop since memories and imperial 
Achaemenid traditions were strong. An inscription in Aramaic letters of not later 
than the third century B.C. on the tomb of Darius at Naqsh-e Rustam has caused much 
controversy, because it is preserved in an extremely miserable condition, and many 
fanciful readings of words based on the line drawings made by E. Herzfeld have 
obscured rather than aided an understanding of the inscription. 56 If the inscription 
does have (or rather had) the name slwk or 'Seleucus' in the second word of the fourth 
line as Henning proposed, then the inscription should be dated probably in his reign 
rather than at the end of the Achaemenid period. 57 The reason for such an unusual or 
unique inscription in a form of Old Persian but in the Aramaic script escapes us, but it 
might be explained as an attempt of local lords to preserve some memory of the 
Achaemenid kings. If Seleucus himself had ordered such an inscription, one would 
have expected a Greek version or an inscription only in Greek, as those from parts of 
altars or bases of statues from a temple in the plain at Persepolis dating from the 
Seleucid period. 58 If slwk is not the name of Seleucus 1, then the inscription might date 

55 Cf. E. Schmidt, op. cil., 2 [ch. 5, n. 141], 1 10- destroying some vestiges of letters on the friable 
1 4 ; G. C. Miles, Excavation Coins from the Persepolis surface of the stone, it is not possible to control his 
Agion.NNM, 143(1959), 1,19; D.Stronach, op. sketches. Nonetheless, an examination of the 
cit. [ch. 5, n. 1], 198. Diodorus (XIX, 48, 5) surface and photographs reveals the inadvisability 
mentions a 'near revolt' of the Persians when of relying implicitly on his drawings. 
Antigonos replaced Peukestas as satrap of Persis, -" W B. Henning, "Mittehramsch," in HO. 
and Thespios their leader was executed by Iranistik (Leiden, 1958), 24. Traces on the stone are 
Antigonos. Later Diodorus (XIX, 92, 5) says that not now observable, so one must accept the 
the troops loyal to Antigonos went over to reading of Henning, the most remarkable epigra- 
Seleucus, the last we hear of Persis in the sources. phist in Iranian studies. The reading saruka 

56 Published by E. Herzfeld, Ahpersische In- 'princeps for slwk by G. Ito, "Gathica," Orient, 12 
sc/ir./ren (Berlin, 1939), 1 2, Tafel I V ; the interpre- (Tokyo, 1976), 58, is intriguing, although the 
tations by F. Altheim, Weltgeschichte Asiens im word would mean rather 'fortress' or 'tower'; cf. 
griechischen Zeitalter, 1 (Halle, 1947), 37-39, Frye, The 'Aramaic' inscription on the Tomb of 
repeated in his Die aranutische Sprache unter den Darius,"//!, 17 (1982), 85-90. 
Achaimeniden (Frankfurt/M, 1963), 10-12, and 58 E Herzfeld, An Archaeological History of Iran 
elsewhere, are dubious. Since Herzfeld made a (London, 1935), 44 He says the inscriptions were 
paper squeeze of the inscription, in the process not from a Greek temple but a temple to the local 



Alexander the Great and the Seleucids \5$ 

from late Achaemenid times, which would make more historical sense than a Seleucid 
date. 

The coins of the frataraka 'governor' of Persis are also enigmatic at least in their 
dating and significance. Generalizations from the presence or absence of coins at 
excavation sites are always tentative, but they may indicate a certain historical 
situation, so an examination of coins found in excavations at Persepolis, Pasargadae, 
Qasr-e Abu Nasr, and Susa can give us some information. No coins of Persis have 
been found at Susa or at Pasargadae, which does indicate that the realm of circulation 
of the coins was small. The two small hoards from Pasargadae have coins of Seleucus I 
but no later Seleucids, while the coins from Persepolis have only one Seleucid 
example but many frataraka coins. 59 This would further imply that Pasargadae rather 
than Persepolis was the Seleucid center in Persis and that Persepolis may have been a 
local center, the two existing in some sort of modus vivendi until the local dynasts 
established a treaty or vassal relationship with the Seleucids. From the coins found in 
excavations in Fars province, and from the bazaars of Shiraz, it seems that the only 
Seleucid coins current in this province were those of Seleucus I, and one small hoard 
from the plain near Persepolis indicates that the coins of the local dynasts in Persis 
followed immediately after those of Seleucus I. 60 The local coins were inscribed with 
Aramaic legends, unlike other coins in this period which were all in Greek, and the 
Persis coins show a development over a long period of time in several phases. The 
earliest coins have the legend in Aramaic prtrk ' zy 7/iy "governor of the gods,' or on 
'behalf of the gods.' The title 'frataraka' is enigmatic, but it is attested in Aramaic 
papyri found in Egypt from the Achaemenid period, and the office in Egypt was just 
under the satrap and included military as well as civil jurisdictions. The etymology of 
the word is uncertain, but it may mean 'the first' (after the satrap) and thus be 
appropriate for a similar office in Persis. 61 We do not know whether the first coins 

gods, although reasons are not given. The short 60 E. T. Newell, The Coinage of the Eastern 

inscriptions are simply the following: HAIOY; Seleucid Mints (N.Y., 1938), 159-60. As far as I 

A BHNAE BAZIAEIAE; AIOE MEHE- know the only Seleucid coins found in the bazaars 

TOY; APTEMIAOE; and ATIOAlAOl all are Alexander type coins and rarely Seleucus I. The 

on small slabs of stone. There is no indication of two strange gold coins supposedly from the Oxus 

assimilation with local deities or syncretism in treasure, now in the British Museum, may be 

these inscriptions. forgeries; at least no. 4 pi. XXVIII of Hill, op. cit. 

59 Cf. G. K.Jenkins ajWStronach, op. cit. [ch. 5, [ch. 5, n. 104] does not inspire confidence in its 

n. 1], 185-98; G. Miles, Excavation Coins from the authenticity. Coin no. 6, obv. has a head like those 

Persepolis Region, NNM, 143 (1959), 1-23, with on the early coins of Persis while the rev. has the 

further references on page one. At Qasr-e Aba quadriga of coins of Andragoras. No other 

Nasr (Old Shiraz) one bronze coin of Seleucus I specimens of these coins have been found, so they 

was found and a copy of a hemidrachm of cannot be used as well-known coin issues. 

Alexander the Great, as well as eleven coins of 6I For a discussion of the title and bibliography 

Persis, cf. G. Miles apud R. N. Frye, Sasanian see P. Naster, "Note depigraphie monetaire de 

Remains from Qasr-i Abu Nasr (Cambridge, Mass., Perside," IA, 8 (1968), 74-80. It is important to 

1973), 26. The existence of a mint in Persepolis or note that one and the same man is called segan (in 

Pasargadae is by no means clear, since those coins Aramaic 'governor") and frataraka in the Elephan- 

attributed to a mint in Persis may in fact be issues tine documents; cf. B. Porten, Archives from 

of Susa. Cf. A. Houghton, "Notes on the Early Elephantine (U. of California Press, 1968), 48, and 

Seleucid Victory Coinage of 'Persepolis'," Schtvei- Grelot, Documents aram. d'Egypte (Paris, 1972), 75. 
zerische Numismatische Rundschau, 59 (Bern, 1 980), 
4. 



160 Chapter VI 

were struck by independent princes or simply vassals of the Seleucids, perhaps similar 
to the family of Atropates in Azerbaijan, but if we follow the evidence of the coins 
alone we could say that after Seleucus I, in Persis we find only local dynasts who issued 
coins with Aramaic legends, and on the reverses of the coins was a structure with 
either the horns which existed on the parapet of the palace of Achaemenid Persepolis 
above the plains, or a crenellated stone, also found on the top of buildings at Persepolis 
above which was the 'Ahura Mazda' symbol (or the king's 'glory"), and beside the 
structure an Achaemenid (war) standard. All of these symbols indicated a 
continuation of the royal and religious traditions of the great kings in their homeland. 
The numismatic evidence seems clear, but unfortunately it does not tally with literary 
sources. 

Polybius is the main source, and he tells of Alexander a Seleucid governor of Persis 
at the accession of Antiochus III in 223 B.C. (V, 40, 7). After the suppression of a revolt 
by Molon, satrap of Media and brother of Alexander, the latter committed suicide, 
and later Antiochus III, on his way back from India c. 205 B.C., stopped in Kerman, 
but nothing is said of Persis (XI, 39, 13). Most scholars have considered this sufficient 
evidence to propose the continuation of Seleucid rule in Persis at least until the end of 
the reign of Antiochus III. Before the eastern expedition, in the battle of Raphia of 
Antiochus III against Ptolemy V of Egypt (c. 217 B.C.) Persian bowmen served in the 
Seleucid army (V, 79, 6). Two notices in Polyaenus, Strategika (7, 39 and 40), refer to 
Persis but with no dates. In the first a certain Seleucid general Seiles under one of the 
kings called Seleucus massacred 3000 rebellious Persians, and later a certain "Ofioptfls 
led the Persians to a massacre of 3000 colonists (k&toikoi), and this latter event has 
been identified with a Vohubarz, second frataraka of the series of Persis coins. Another 
event, mentioned by Pliny (VI, 152) is the victory of Numenius, the governor of 
Mesene under Antiochus, over the Persians in a sea battle in the straits of Hormuz off" 
the coast of Kerman, but we do not know which Antiochus is meant, in this case 
possibly IV. All of this has led some scholars to propose that Persis only became 
independent after Antiochus III, with the two events mentioned by Polyaenus and 
Pliny's remark coming later. 62 How do we reconcile the numismatic data with the 
statements by Classical authors? There are two solutions; either Seleucid control rose 
and fell from the time of Seleucus I, which might fit with the different series of coins 
of Persis, or there may have been a parallel rule, either in different geographical parts 
of Persis, the Seleucids in the north and Persians in the south, or a concurrence of 
Seleucid and Persian rule where a Seleucid satrap ruled together with a local dynast. 
The title of the Persian dynast could imply a division of authority between a Seleucid 
civil governor and a religious, Oriental local ruler who, nonetheless, had the right to 
strike coins for the province. The absence of Seleucid coins in Persis, while in other 
parts of Iran they are found, mitigates against accepting a full Seleucid control with 
the dating of the rule of the local dynasts only after Antiochus III. 63 It is possible that 

62 H. H. Schmitt, Untersuchungen zur Geschichte B.C. This is hardly acceptable, for the coins must 

Antiochos' des Grolien und seiner Zeil, Historia, cover a longer period of time. 
Einzelschriften, Heft 6 (Wiesbaden, 1964), 46-50, 63 On Seleucid copper coins found in Media see 

and esp. Stiehl in Altheim, Huntien, 1 , [ch. 5, n. 68], M. A. Stein, Old Routes of Western Iran (London, 

379, who puts the rule of the frataraka dynasty in 1940), 304-04, and Newell, op. cit. [n. 60], under 

Persis less than 50 years, between c. 1 87 and 1 40 'Ecbatana.' 



Alexander the Great and the Seleucids 161 

the Seleucid kings, at least from Antiochus I through III, had an agreement with the 
local dynasts of Persis, allowing them to strike their own coinage, and to give only 
nominal allegiance to the Seleucids, who nonetheless did maintain a presence in the 
person of a satrap appointed by the king. Since the province was very difficult of 
access because of mountains on all sides, and not lying on important trade routes, this 
picture may be accurate. Unfortunately, we do not know the locations of Seleucid 
city foundations in Persis, but Tarn has argued that the most important, Antiochia, 
was on the coast near present Bushire. 64 The remark by Strabo (XV, 736) that the 
Persians in his time had kings subject to other kings, formerly to those of the 
Macedonians but now to the Parthian (kings), perhaps provides the answer, namely 
that the dynasts of Persis who issued their own coins were really vassals of the 
Seleucids. 

The names of the early rulers of Persis were not Achaemenid names; those came 
much later in the series which we will examine when the Persis dynasts were under 
Parthian hegemony. The early coins show on the obverse a head facing to the right 
with a distinctive headdress covering ears and hair, unlike any previous royal 
headgear but similar to heads on some satrapal issues of the Achaemenid period, 
which may be an indication of the status in which the dynasts regarded themselves. 65 
The legends on the coins are extremely difficult to decipher and many letters cannot 
be distinguished from each other, for on several specimens of the earliest coins from 
the series, one scholar read bywrt, while another read bgkrt (the latter more likely as an 
Iranian name) with his successor bgdt 'god given.' 66 However the dispute over the 
date of these coins may be resolved, it seems clear that the coins of the first series, using 
the title frataraka, imply the subordination of a dynast to a higher sovereign, and 
whether this first happened after Seleucus I or Antiochus III is at present unknown. 
The second series of these coins show a change in headgear and a degeneration of the 
reverses, but more important is the change in title to MLK' 'king,' and old 
Achaemenid names such as Darius (d\yw) and Artaxerxes (\tx$tr) appear on the 
coins, while the final series takes on the characteristics of Parthian coins. It is clear that 
after the reign of Seleucus I, Persis was not securely under Seleucid rule, but a full 
independence of the province seems possible only after Antiochus III. If the coins of 
Persis were struck only after that time we must suppose more than a century of no 
coinage in the province after Seleucus I, and a corollary would be that only the Greco- 
Macedonian settlers used the coins of Seleucus I while the local population did not, 
and only later did they accept their own local coinage, but this is unlikely. It is safe to 
assume that in the province of Persis old imperial memories, rather than traditions, 



64 Tarn, Alexander, 1 [n. 6], 257; Greeks [ch. 1, 66 The first reading was by Hill, op. cit. [ch. 5, n. 
n. 23], 418. This is a plausible suggestion since, as 104], clxv and 196; the second by Allotte de la 
we shall see later, Antiochus HI probably launched Fuye, Mordtmann and others; cf. de Morgan, 
his expedition against Arabia from this city - a NuiiiismatiqueJelaPerseanlique(Pms,1927), i55- 
seaport. 79. On some coins we seem to find prs prtrk ' 'Persis 

65 de Morgan, Manuel, 1 [ch. 5, n. 103], 50, fig. frataraka.' 
29. See also P. Naster, "Fire-altar or Fire-tower on 

the Coins of Persis," Orientalia Lovaniensis, 1 
(1970), 125-29. 



162 Chapter VI 

were preserved but romanticized and distorted, whereas elsewhere in Iran the past did 
not weigh heavily, if at all, on the local population. 67 

Kerman or Carmania was an eastward extension of Persis, although it was 
separated from Persis by deserts and mountains and did not have the isolation which 
Persis did. Communications with Media were easier than with Persis, and when we 
hear that Antiochus III in the fall of 206 B.C. reached Kerman coming from his eastern 
campaign (Polybius XI, 39, 13), we are not told how he returned to Seleucia, nor do 
we hear of a Seleucid satrap in Persis at this time, then one can only speculate on the 
relation of Antiochus III to the local population. Pliny (VI, 115) says an Antiochus 
built a town to the west of Pasargadae, which Antiochus we cannot say, but probably 
II, while Antioch in Persis was built by Antiochus III about 205 B.C. and, as noted, has 
been located near present Bushire. 68 If this location is correct, it tells us nothing about 
Seleucid control in the interior of Persis, for the sea coasts were surely under Seleucid 
rule for a long time. It must be remembered that Antiochus III lost his life while 
attempting to plunder a temple in Elymais in the summer of 1 87 B.C. and by that time 
Seleucid hegemony in Persis and Elymais must have been most tenuous if existing at 
all. 69 

To turn to the north of Iran we find more traces of the Seleucids, both in 
monuments and in coinage, than in Persis. The two stone columns and stone bases 
possibly of a temple at Khurhe (near Mahallat) date from Seleucid times, as does the 
late relief of the reclining Herakles from Behistun. 70 In 193 B.C. Antiochus III 
instituted a cult for his wife Laodice as recorded in a Greek inscription from 
Nihavend, and another inscription on the relief of Herakles dates from 148 B.C. Other 
remains of the Seleucid occupation of northern Iran are few, but so are the non- 
Hellenistic remains such as the rock tombs of Kizkapan in Iraqi Kurdistan, which 
show as much Hellenistic influence as the frataraka figures on the door jamb from 
Persepolis. Farther to the north, in Azerbaijan and in the Caspian provinces both 
Hellenistic cultural influences and political control were much less than in Media. 

All indications point to Hamadan as the summer residence, and Seleucia as the 
winter capital, of the Seleucid co-ruler, the governor-general of the 'upper satrapies,' 
who was the crown prince Antiochus under Seleucus I beginning from the year 294 
or 293 B.C. We do not know whether succeeding crown princes automatically 
occupied this high post in virtue of their being successors to the throne, but we may 
assume several of them did. 71 Antiochus III, we have seen, assigned Molon to this 

The only certain archaeological remains town but had been driven out may be compared 

from this period in Persis are the hardly visible with I Maccabees 5, where the town is called 

figures on a door jamb of a temple near Persepolis; Elymais! Whatever the reality behind these 

cf. E. Schmidt, op . cit., 1 [ch. 5, n. 141], figs., 16-17; reports, they show the loss of Seleucid rule there 

R. Ghirshman, Iran, Partitions, and Sassanians 70 All bibliographic references to archaeological 

(London, 1962), 26, fig. 34. sites and monuments may be found in L. Vanden 

For the text see C. B. Welles, Royal Berghe, Bibliographie analytique de I'archMogie de 

Correspondence of the Hellenistic Period, (New I'lran ancien (Leiden, 1979), and references will be 

Haven,1934)31/2.Tarn,Greefej[ch.l,n.23],418, made to this work. Seleucid copper coins have 

for identification of this town with Bushire; see been found near Harsin and elsewhere in Media; 

also H. H. Schmitt, op. cit. [n. 62], 14. see Stein, op. cit. [n. 63], 304-07. 

69 The report in II Maccabees, 9, 1-2, that 7I Bengtson, Strategic 2 [n. 46], 83-85, gives a 

Antiochus IV had entered Persepolis and tried to survey with references, 
rob the temples there and take possession of the 



Alexander the Great and the Seleucids 163 

position, and later another rebel against the Seleucids, called Timarkhos, was also 
probably the governor-general of the 'upper satrapies.' As time progressed, however, 
the area of the 'upper satrapies' diminished, and we cannot determine the extent of 
Seleucid rule in each of the provinces in the east. 

Azerbaijan, or Atropatene, named after the satrap Atropates, was no more part of 
the Seleucid realm than Persis, and the same lack of information about a Seleucid 
satrap, relationship to Media or vassal status, makes any precise determination of rule 
impossible, but the area was hardly under direct central control of the Seleucids 
(Strabo XIII, 523). The summer capital of Azerbaijan was Gazaka or Ganzaka 'the 
treasury' probably located near modern Miandoab, while the winter capital of 
Phraaspa or Fraata was near modern Maragheh. 72 When Antiochus III invaded the 
area and then made peace, the local dynast who continued to rule was called 
Artabazanes, a member of the same family as Atropates. The isolated Caspian Sea 
coastal area, as frequently throughout its history, was independent of control from the 
plateau. 

THE FALL OF THE SELEUCIDS AND THEIR HERITAGE 

From the coins minted in the eastern part of the Seleucid empire, one tentatively may 
trace the decline of Seleucid control on the Iranian plateau. If the observations about 
the coins by Newell are accepted, Antiochus first struck coins in the chief city of 
Bactria after he had been made ruler of the eastern satrapies, and the coins should date 
after 289 B.C. 73 The change in weight standard from Attic to 'Indian' in a second series 
of such coins does not mean that the coins were struck in India but rather in an area of 
eastern standard weight in which India was included. The series of Seleucid coins 
struck in Bactria ends with Antiochus II and naturally blend into issues of the first 
independent ruler, Diodotus, first with the name of Antiochus and then his own 
name, indicating the stages of transition to independence probably after the death of 
Antiochus II in 247 B.C. (Newell, p. 249). To conclude questions of coinage, one 
group of enigmatic coins does not fit into any series we know; the name in Greek on 
the obverse of the coins showing a bust of someone in a helmet copied from that of 
Seleucus I, without a title, is in the genitive, Uaxpvrov, with weight based on an 
Indian rather than Attic standard. 74 In spite of the frequent identification of the name 
on the coin with an Indian ruler Sopeithes, a contemporary of Alexander the Great, 
this is hardly possible from a numismatic point of view. The mint should be in 
Bactria, or possibly at Kapisa in the Hindukush, but the dynast or satrap on the coin 
from the style of the coin should have lived about the time of Seleucus I or later. His 
coins may be compared with those of Andragoras, satrap of Parthia who became 
independent in the time of Seleucus II. 

Uncertain mints in the east, unfortunately, cannot help us with a chronology of 
Seleucid rule in various provinces such as Aria (Herat), Seistan or others, mainly 

72 See V. Minorsky, "Roman and Byzantine R. B. Whitehead, "The Eastern Satrap Sophytes," 
Campaigns in Atropatene," BSOAS, 11 (1943), Numismatic Chronicle (1943), 60-72. See also C. 
263. Kirkpatrick, "Some New Coins of Sophytes," 

73 Newell, op. cit. [n. 60], 230. Numismatic Circular (London, Oct. 1973), 373, 

74 All the literature on Sophytes is collected by where 19 coins are listed. 



164 Chapter VI 

because the mint sites cannot be located. From all the coins, however, it is clear that 
Seleucid rule came to an end in eastern Iran more or less with the death of Antiochus 
II, if not before, and only the expedition of Antiochus III to the east briefly restored 
Seleucid rule. It is also evident that the mints in Hamadan and Bactra were the two 
important, continuous mint sites in the east, an indication of Seleucid interest in 
controlling the trade route from the plains of Mesopotamia to Central Asia. 75 The 
existence of coins with the names Sophytes, Andragoras, and in Aramaic wMw{r) 
'Vaxsu' or 'Vaxsuvar,' without titles indicates a gradual process of the loosening of ties 
with the Seleucid court. 76 The existence of this rare Aramaic legend on the coins 
raises the question of the position of Aramaic in the Seleucid realm. 

Greek was the official and dominant written language of the Seleucid empire, but 
Aramaic continued in use especially in the outlying provinces or principalities such as 
Armenia, Sogdiana, Khwarazm, or Persis, while Asoka, the Mauryan ruler (c. 273- 
232 B.C.), as we have seen, considered Greek and Aramaic the two languages of the 
inhabitants of the Qandahar region, although in Laghman, east of Kabul, only 
Aramaic was deemed sufficient to record his Buddhist edicts. Did the Seleucids have a 
dual chancellery, using both Greek and Aramaic as 'official' languages? There is no 
evidence, other than bilingual inscriptions, for the use of Aramaic by the Seleucid 
government except to communicate with the literate local population or with 
independent or semi-independent satraps or dynasts. We may conjecture that Greeks 
and Macedonians living in a polis or in a settlement in the east at first followed the 
usual pattern of ruling colonizers; they did not learn the local language as much as the 
local population had to learn Greek. Since most of the Greek colonies in the east were 
set up near native towns or villages, and Seleucid cities seemingly were not so clearly 
divided between colonists and natives as in Egypt, this made the process of 
Hellenization, and the opposite as well, progress more quickly than in the Ptolemaic 
kingdom. 77 Obviously this is not the place to discuss Seleucid institutions, for even 
information about military colonists (katoikoi), land grants (kleroi), or ethnic groups 
in a town (politeumata) comes from the western part of the Seleucid realm, and it is 
hazardous to transfer a situation in Anatolia to one in Merv, for example. 78 Certainly 
in Bactria military colonies were numerous, composed of retired soldiers and their 
families, as well as civilian colonists brought from the west and given land by the 
government. The military colonial origin of the majority of city foundations in the 
east is clear, but we have no information about how different groups of people lived 
in them and how they reacted together. How many Greeks and how many Orientals 
lived in one or another colony escapes us, as do the reasons why some became poleis 



75 That Seleucus founded cities along trade now in the Bibliotheque Nationale in Paris, 
routes is stated by Libanius, Oral. 11, 1 00. Cf. indicates a mint master rather than the name of a 
G. M. Cohen, The Seleucid Colonies, Hisloria, city. Cf. Houghton op. cit. [n. 59]. 
Einzelschriften, 30 (Wiesbaden, 1978), 83-84. " Cohen, op. cit. [n. 75], 52-54, 72-83. He 
Strategic considerations were also important in the characterizes Seleucid cities (p. 41) as "Greek in 
founding of colonies. character but distinctly cosmopolitan in 

76 On the coins with Aramaic legends see A. R. population." 

Bellinger, "Coins from the Treasure of the Oxus," 78 One must neither use ancient Greek political 

ANSMN, 10 (1962), 67. The existence of two theory for the Hellenistic age, nor confuse the 

Aramaic letters on a 'victory coin' of Seleucus I, western part of the Seleucid realm with the east. 



Alexander the Great and the Seleucids J65 

while others did not. One may attribute loyalty to the Seleucids not only to the 
feeling on the part of the Greco-Macedonian colonists that it was to their interest, 
indeed their need to survive, to support the house of Seleucus, but also to the policies 
of conciliation and even syncretism in religion, and in customs, which brought 
support from the local population. Undoubtedly Greek legal practices, institutions of 
city life and much more, had a lasting influence on the local peoples under Seleucid 
rule, but to trace or disentangle the connections is hardly possible. 

The Seleucids had successfully assumed the imperial mantle from Alexander and 
the Achaemenids, and the Iranians rather than supporting this, merely accepted it 
while turning their attention to local affairs. There was no imperial pretender around 
whom all Iranians would rally to oust the conquerors. Furthermore, since the Greco- 
Macedonians were tolerant and did not disrupt native religions but rather fostered 
them in the past tradition, they did not encounter any popular movements to 
overthrow the Seleucids but only local or tribal revolts which did not pretend to 
imperial claims of power. The danger to Seleucid rule did not come from these threats 
from natives, but rather from its own satraps or dynasts when they felt that the 
Seleucid central government was weak, and it was the time for them to assert claims 
of independence. The central authority was upheld essentially by a mercenary 
professional army, which when well paid was more than a match for local 
contingents. Inasmuch as the rule of the Seleucids was highly personal the court and 
the 'friends' of the king were very important. Just how much of the court in Antioch 
was mirrored in the court of the governor-general of the east is difficult to determine, 
but one may assume that under Seleucus I, when his son Antiochus was ruler of the 
east (c. 294-281 B.C.) the court of the east was a copy of that in the west, but 
afterwards we have no information even to hazard a guess. One may only conjecture 
that after or even during the reign of Antiochus I the east lost any parity it may have 
had, for the energies of the later Seleucid kings were directed to affairs related to the 
Ptolemies and to Asia Minor, while the east fended for itself. Under the governor- 
general of the 'upper satrapies,' how was Iran ruled? 

We cannot determine whether the political theory whereby the Seleucid realm 
was divided into vassal kings (fiaoiAels), vassal dynasts (Bwdarai), cities and peoples 
(e-frvrf) had any reality in Iran, but all categories existed there. 79 Under the kings we 
could place frontier lords, who were really independent and only paying lip service to 
Seleucid overlordship, or even that being fictional, such as Chandragupta under 
Seleucus I, the kings of Khwarazm in Central Asia and possibly the rulers of Armenia. 
Under the second category one might put the frataraka of Persis and later Andragoras 
and even Diodotus of Bactria, but this is simply conjecture, the need to find someone 
to fit into the second group. Cities in the east are a problem since we only have 
archaeological evidence for them, but it would not be amiss to suggest that they had 
less autonomy than those of Asia Minor, or Seleucia on the Tigris. Although some of 
these cities during the later Seleucid period were almost independent and in the status 

79 On the theory see E. Meyer, Bliite und describing different elements of power and author- 

Niedergang des Heilenismus (Berlin, 1925), 43, and ity, and may have had little or no application to 

Rostovtzeff, op. of., 3 [n. 46], 1440, n. 277. This reality. In the east we simply do not know, 
theory may only have been a general way of 



166 Chapter VI 

of allies of the king, it is doubtful if any of the cities on the Iranian plateau were 
independent of the satraps in the areas where they were located. 80 There is no reason 
to believe that Ay Khanum on the Oxus, for example, had the same relation to the 
local satrap as some of the allied cities of Asia Minor had to later Seleucid rulers. This 
does not mean that the cities in the east had no internal autonomy; they probably 
collected their own taxes to remit to a central government, and they had their own 
Greek forms of government. Just how the king's or satrap's representatives meshed or 
clashed with local authorities is unknown. 

The relationship of land tenure and land ownership to cities and other institutions 
of the central government also must be surmised. In addition to old Achaemenid 
'royal' domains which were given to Greek colonists, retired soldiers, or left in the 
hand of an original Iranian landlord, some lands were granted by the king directly to 
a city or a colony. Temple property in the early years seems to have been little 
touched by the king, although later in the Seleucid period sequestering of temple 
property by the government or king was not uncommon. There is also no evidence 
of attempted 'Hellenization' by force in Iran, as under later Seleucid kings in Palestine/^ 
and if sources from Babylonia provide a model for non-existent sources in Iran, then 
we may infer a Seleucid tolerance, if not patronage, of temples and religions in the 
east. 81 The existence of rich and powerful temple complexes in Iran, as existed 
in Babylonia, is unattested, although shrines and sanctuaries existed, and the 
architectural remains of what may have been temples in Khurhe and Kangavar show 
Greek styles and influences. Were such structures built by and meant only for the 
Greco-Macedonians, or did Iranians participate in the worship of Herakles, Dionysius 
and other Greek deities? What seems to have happened was the adoption of certain 
Hellenic traits, stories, or even practices in the religion of the Greeks, by Iranians into 
the Zoroastrian religion without, however, direct 'conversion' or even the adoption 
of foreign names. Thus Rustam, probably originally a Saka hero later adopted by all 
Iranians, took on some of the traits of Herakles without becoming the Greek hero in 
name or religious content. It is, of course, misleading to speak of one 'Greek' religion 
or one 'Iranian' religion, since there undoubtedly were many local cults and variant 
forms of religious rituals and practices among both peoples. Since both peoples were 
heirs of a common Indo-European religious, as well as linguistic, background such 
identifications as Helios or Apollo with Mithra and Herakles with Verethragna at 
Commagene in Asia Minor must have existed elsewhere. Just how much 
Zoroastrianism clashed with the Greek religion, but especially with the Hellenistic 
ruler cults, we cannot say. Whether Iranians distinguished between Greeks and 
Macedonians, or Hellenized Asiatics is also unknown, but unlikely. In fact, our lack of 
information from Iran in this period is especially frustrating, because we must rely on 
Babylonia or Bactria to give possible parallels. 

In Babylonia Seleucia was a Greek polis while Uruk was a native city, and 

80 Certain cities, such as Seleucia on the Tigris, s ' Sec Rostovtzeff, op. cil , 3 [n 46], 1 427. There 

o\erstruck imperial coins indicating some kind of is no evidence that the Seleucids killed native 

independence, and if the cities did this we may priests and destroyed temples, and if it happened 

infer that satraps in the east did the same, in the case the motivation was political rather than religious, 

or Sophytes, Andragoras and others Cf. Bikerman as far as we know. 
op (it |n. 46], 226 



Alexander the Great and the Seleucids 167 

excavations have been made at both. 82 It is clear that at both cities Seleucid coins were 
current; Greek inscriptions on seals and bullae are found; Greek officials were present 
in both, and other features of Hellenism were present in both. But Uruk had a temple 
to the God Anu, identified with Zeus, and all of the cuneiform documentation shows 
a traditional native life in full vigor, while Seleucia was founded as a Greek city 
without local elements. Ay Khanum was apparently like Seleucia, but in time all of 
the towns became mixed in character if not in institutions. Some natives were proud 
to become 'Hellenized,' others were not. Some aristocratic Iranians became 
'Hellenized,' as did their descendants who became 'Westernized' in recent times. In the 
end native culture and values absorbed the 'Hellenization,' but it took a long time and 
the tenacity of Greek culture was remarkable, especially the Greek language and 
institutions, for the Greek cities continued to exist late into Parthian times still 
holding on to civic organizations, gymnasia, theatres and such features of Hellenistic 
civilization which did not interfere with the religious and domestic life of the mass of 
the population. For native customs and traditions prevailed in time over the laws and 
institutions grafted onto the ancient lands by the Seleucid kings. The peasantry, the 
vast majority of the population, continued to exist, little changed by the Hellenism of 
the cities. We do not know of conflicts between the countryside and the cities, not to 
mention the role of nomads, but one may assume that under the Seleucids there was 
less of this conflict between different groups of the population than in other periods of 
Iran's history, although such conflicts may well have contributed to the fall of the 
Seleucids. 

Little has been found about monetary policies of the Seleucids, but again, by 
comparison with other areas such as Babylonia and Syria, one may project into Iran a 
picture of a great expansion of trade and commerce on the Iranian plateau with the 
Seleucid policy of founding cities and continuing coinage started by Alexander on the 
Attic standard. Most trade in the east was in luxury goods, such as spices, gems, and 
textiles, for the east was largely self-sufficient in foodstuffs, unlike Greece, but no 
doubt the extension of the use of coinage into areas which previously had known only 
barter and the expansion of contacts with foreign lands greatly influenced Iran and 
Central Asia as it did the rest of the Hellenistic world. The introduction of the 
Seleucid calendar to all parts of the empire, rules and regulations on banking, and 
many practices of which we have virtually no information, all indicate that the early 
Seleucid period was one of organization and development, expansion of trade, and in 
general of great prosperity. 83 But it did not last, mainly because of the continual 
warfare between the Seleucids and the Ptolemies and between the Seleucids and 



82 For the various reports on Uruk (modern rir ,2 |n. 46], ch. 8. It is impossible to do more than 
Warka) see the references in B. Hrouda, Vordera- touch the many activities in the Seleucid period 
sien I, Handbuch der Archaologie (Munich, 1971), which brought Greece and India in contact, not to 
292-93 On Seleucia sec W van lngen. Figurines mention countries in between. Travelling artists, 
from Seleucia on the Tigris (Ann Arbor, Michigan, musicians and players as well as merchants moved 
1939) with an extensive bibliography. On the cults great distances in the Seleucid period. Cf. The 
it Uruk see M Rutten, Contrats de Vepoque l:\cavations at Dura-Europos, Ninth Season, pre- 
Seleucide (Paris, 1935), 28, 37 and 52, for an hminary report, ed. by M. I. Rostovtzeff, 1 (New 
assimilated cult of the king. Haven, 1944), 264. 

83 See the excellent summary in Rostovtzeff op 



168 Chapter VI 

various Hellenistic states of Greece and Asia Minor. It began already in the time of 
Antiochus I. 

Antiochus I, after the death of his father, probably never returned to Iran, for he 
had to fight many battles especially in Asia Minor, revolts in Syria and wars with the 
Ptolemies, although in the decade before his death (271-261 B.C.), he may have gone 
to the east after Babylon where he was in 268 B.C., during the crisis when Antiochus 
had his eldest son Seleucus killed, and a younger son Antiochus II became crown 
prince. The latter, after the death of his father, had to fight much and often against the 
Egyptians and others, while it seems that affairs in the east were neglected even more 
by Antiochus II than by his predecessors. The satrapies of the east were left to fend for 
themselves, and one may conjecture that sentiments for independence from central 
Seleucid authority were not slow to appear and by 246 B.C., the death of Antiochus II, 
de facto independence was moving to complete separation. Under Seleucus II, who 
had to fight for his throne in the west against his half brother supported by the 
redoubtable Ptolemy III, the eastern part of the Seleucid realm began to disintegrate. 

It is significant that in the mint of Bactra the coins of Antiochus II are followed bf 
coins with his name but with the head of Diodotus, presumably satrap of Bactria, and 
finally the same type coins with the legend BAZIAEQE AlOAOTOY. 84 This 
implies that the break with the Seleucids took place after the death of Antiochus II in 
the reign of Seleucus II. Just who was governor-general of the eastern provinces at this 
time, or whether the office had lapsed, we do not know. About the same time we may 
suppose, a certain Andragoras, usually identified as satrap of Parthia, issued coins 
without a title, indicating a continuing nominal allegiance to the Seleucids. Since a 
Greek inscription with the same name, not a satrap but a lesser officer, has been found 
in Hyrcania (Gurgan), but from an earlier period before 266 B.C., it is possible that we 
have the same person, who later became satrap of Hyrcania or of Parthia. It is also 
possible that the coins with the name Sophytes, without title mentioned above, also 
are from this period, with Sophytes as a lesser satrap somewhere in the east. This 
would mean that the eastern satrapies as a group reacted against Seleucid domination 
and central weakness about the same time, the beginning of the rule of Seleucus II. 
While this is only a hypothesis, it covers our known data better than any other 
surmise and may stand until further evidence invalidates it. 

Seleucus II was in no position to move eastward to recover lost Seleucid lands until 
he was driven from Asia Minor by a great defeat at Ankara, probably c. 239 B.C., but 
then he had to reorganize his forces in Syria and Mesopotamia. So his campaign to the 
east must have been sometime, as a guess, between 230 and 227 B.C. 85 The inability to 
assign coins of Seleucus II to a definite mint in eastern Iran (Newell vacillates between 
Hekatompylos and Herat), combined with the paucity of sources (Strabo XI, 514, and 
Justin XLI, 4, 8-10; 5, 1) prevents us from reconstructing events in the east. Strabo 
says the Parthians fled before Seleucus into the desert and Justin claims the Parthians 

84 On the Bactrmn coins see Newell, op cit. (n 85 On the dating and a good discusion of this 

60], 247-49; on the coins of Andragoras see R. period, see Will, supra, Histoire politique, 1,278-81, 

Ghirshman, "Jhe tetradrachmc d"Andragoras," in with an ample bibliography. 
Near Eastern Numismatics, Studies in Honor of 
George C. Miles, ed. by D. K Kouymjian (Beirut, 
1974), 1-8 



Alexander the Great and the Seleucids 169 

defeated him. In any case, Seleucus II was not able to restore rule over the east, and no 
more Seleucid coins were minted in the east, although Antiochus III did strike coins at 
Hamadan, the last vestige of a continuing Seleucid presence in Iran. 

After a short rule, Seleucus HI was succeeded by his brother Antiochus III, and the 
fortunes of the Seleucids seemed to revive as do our sources. Antiochus at the death of 
Seleucus III was in the east, although Polybius (V, 40, 5) does not say he was governor- 
general of the upper provinces as many scholars have assumed. The historian 
continues that after becoming king he entrusted the rule of the upper parts to Molon, 
satrap of Media and his brother Alexander in Persis. This implies that only Media 
(minus Azerbaijan) and Persis (probably minus most of the southern part of the 
province including Istakhr-Persepolis) remained under Seleucid control at the 
accession of Antiochus III. We are not concerned with actions at the Seleucid court or 
in the west which may have influenced Molon to revolt, the news of which came to 
Antiochus in the summer of 222 B.C., but only with events in Iran. Molon issued coins 
at Hamadan and then at Seleucia on the Tigris probably in the fall of 222, after Molon 
had defeated two generals sent against him by Antiochus, who was busy in Syria 
fighting Ptolemy IV. The exact time when Molon took the title of king and struck 
coins cannot be determined, but the number of coins which have survived suggests 
that Molon took the title shortly after his revolt, which only lasted two years. Did he 
consider himself almost a co-king, as governor-general of the 'upper satrapies,' or was 
his pretension even greater from a mere satrap to be Seleucid king? We must examine 
the title of governor-general of the upper satrapies' to determine what it was, since 
some scholars claim that there were two governors-general, one in the east and one in 
the west from the beginning of Seleucid times. 86 

There is no evidence for such a post, which we may call a viceroy of the east, in 
Achaemenid or Alexander's time, and the first intimation of such an office is under 
Antigonos, but it is uncertain whether a purely military central command is intended 
or a viceroy. Antiochus was certainly that exalted in the latter part of his father's 
reign, and this is the first attested office of viceroy, and the 'capital of the east,' Seleucia 
on the Tigris, became the seat of Antiochus (Appian, Syr. 62). After this time there is 
no evidence for such a viceroy until the time of Antiochus III when the evidence is 
unclear, but from a passage in Livy (XXXV, 13, 5) that the eldest son of Antiochus III 
about 193 B.C. was sent by his father ad custodiam ultitnarum partium regni, many 
scholars have asserted that the son was made viceroy of the east, but this is again only 
inference. Finally, one of two inscriptions from Nihavend speaks of a certain 
Menedemos as (o) tiri twv avco oarpaireicbv at the end of the reign of Antiochus III 
or more likely under Seleucus IV, his successor. 87 The same title is found later, so one 
may ask whether this particular title, as opposed to the viceroy (compare Antiochus 
in the reign of Seleucus I), was not a creation of the administrative reforms of 
Antiochus III. As Bengtson has suggested, control of each province in the east was put 

86 Bengtson, Griechische Geschichte [n, 46], 439. 87 L. Robert in Helleiiica, 8 (Paris, 1949), 23, and 

According to Bengtson, Strategic [n. 46], 86, and 9 (1950), 73. Another inscription on a rock relief 

Will, op. c it., 2, 1 0-1 1 , Antiochus held the post of of a reclining Herakles reads virep T7js KXtofievov 

governor-general of the east until he delegated the tov ini roiv dvu> oarpaTreiaiv awrripias, which 

post to Molon on becoming king, but there is no gives pause to any inflated conception of such a 

direct evidence, only inference. title at the late date of 148 B.C. 



170 Chapter VI 

under a general (aTparryyos) who took over the functions of a civil governor 
(satrap). 88 A further inference from this is the systematic division of the former 
satrapies into smaller units, although this process may have been begun earlier, and 
became progressive through time. In the east, however, the last great expedition of 
the Seleucids, under Antiochus III, although it for a time restored some allegiance to 
the Seleucids, did not reverse the process of disintegration. 

After the end of Molon's revolt in 220 and the submission of a local Median dynast 
called Artabazanes by Polybius (V, 55, 2), Antiochus III had to return to Syria and the 
west to settle affairs, but about 212 B.C. he began his famous expedition to the east. 
First he went against Armenia, which included present-day northern Iranian 
Azerbaijan, and he was successful in restoring the allegiance of Xerxes the local prince 
of western Armenia (center in Sophene) (Polybius VIII, 23) as well as others, but later 
Antiochus sent two generals, Artaxias (Artases in Armenian) and Zariadris (Zareh in 
Armenian) to rule eastern and western Armenia, both of whom became independent 
after the battle of Magnesia in 189 B.C. Securing tribute from this expedition, 
Antiochus now prepared to emulate Alexander in the east. By 209 he w£s 
campaigning in Parthia, after first stripping a temple in Hamadan of its precious 
metals to help pay for his expedition (Polyb. X, 27, 1 3) . How much one can add to the 
short account of Polybius is uncertain, but it seems that peace was made in which the 
Parthians recognized in some way the suzerainty of Antiochus III, and the road 
eastward was free. Antiochus continued against King Euthydemus of Bactria, who 
first opposed him with a large force on the Arius River, which is probably the present 
Hari Rud, an indication of the westward expansion of the Bactrian kingdom. After 
defeating the Bactrian cavalry Antiochus marched to Bactra and besieged 
Euthydemus in his capital, but after a long siege peace was made, the conditions of 
which, however, are unknown. 89 Then Antiochus crossed the Hindukush 
mountains, made peace with a local Indian potentate and having obtained tribute and 
elephants, returned to the west by Arachosia, Drangiane (Seistan) and Kerman where 
he spent the winter, and Polybius gives no more information about his expedition to 
the east. From the lack of information several scholars have inferred that the three 
provinces mentioned above were ruled by satraps loyal to the Seleucids, and 
Antiochus returned via Persis which was either loyal or forced to reacknowledge 
Seleucid supremacy. 90 There is no evidence at all that Arachosia and Drangiane were 

88 Bengtson, Strategic [n. 46], 2, 1 44-98, and his Media, one can understand how as late as 1 48 B.C. 

Griechische Geschkhte [n. 46], 439. Note that the the inscription from Behistun could still speak of 

concept 'upper countries' (or satrapies, or lands) is 'the upper satrapies,' and mean only a small area to 

Iranian in origin, possibly used first under the the west of Hamadan. 

Achaemenids as Khurasan and farther east, while 89 Polybius XI, 39, says Euthydemus persuaded 

under the Seleucids the concept meant the Iranian Antiochus that both would suffer if nomads took 

plateau From the Iranian viewpoint it is difficult advantage of their quarrel to invade the land, so 

to conceive of 'the upper lands' as including the peace was made and a daughter of Antiochus was 

plains of Mesopotamia, although the Seleucids may given to Demetrius the son of Euthydemus. Will, 

have later thought this. I suspect, however, that the Bengtson and Schmitt have little to add or to 

expression simply meant the chief governor in interpret from Polybius, while Justin is of little 

charge of the satrapies on the plateau, which by the help. 

time of Antiochus III meant simply Media and the 90 Schmitt, Antiochus III, 82; Will, supra, 

northern lowlands of Persis. As the new satrapies Histoire politique, 1, 53. It is, of course, quite 

became subdivisions of the former satrapy of possible that thefrataraka ruler of Persis was treated 



Alexander the Great and the Seleucids 1 71 

ruled by Seleucid satraps, but northwestern Kerman, as an extension of Media, might 
well have had a Seleucid governor. It does not follow that Antiochus returned to the 
west through the heart of Persis, for the easiest route would have been, as it is now, to 
Yazd, Isfahan and to Hamadan, not directly to the west over desert and mountains to 
Persepolis. Although there is no evidence for Antiochus' taking the route from 
Kerman to Hamadan, it is an alternative to the usual assertion that all of Persis was a 
satrapy of the Seleucid Empire at this time and that the coins of the frataraka are all later 
than this time. As mentioned above, I suggest that Persis must be divided between the 
northern lowlands (Yazd- Abadeh of today) and the highlands of Persepolis and the 
south, which could explain the supposed rule of frataraka and Seleucid satrap in one 
province. The fact that at the end of the 'Anabasis' of Antiochus III he is found in a 
town Antiochia in Persis, probably on the Persian Gulf, according to an inscription 
dated 205 B.C. addressed to magnates of the city of Magnesia on the Meander River in 
Asia Minor, only supports the contention that Antiochus was in this site in 
preparation for an expedition against Arabia. 91 In any case, the position of Persis 
cannot have been radically different from other outlying provinces of the Seleucids 
where Antiochus wisely realized he could no longer rule directly with the resources 
he had. He might win battles, but continuous rule would be difficult if at all possible. 
The great expedition to the east undoubtedly did re-establish Seleucid prestige and 
even tribute and allegiance, as Polybius says, but times had changed since the first 
Seleucus. After the defeat of Antiochus III by the Romans at Magnesia in the 
beginning of 1 89 B.C., it is difficult to believe that any of the agreements to pay tribute 
to Antiochus III remained in force in the east save in Media which was always the 
center of Seleucid power and authority for the east. For the Iranians, as for 
Euthydemus and others, the expedition of Antiochus to the east probably was 
regarded as an attempt to obtain booty and to forge alliances which might be of use to 
the Seleucids in the future, but the last was a vain hope. Antiochus III plundered a 
temple at the beginning of his expedition, and to pay reparations to the Romans he 
lost his life in attempting to plunder a temple at Elymais on July 3 or 4 of the year 187 
B.C. 92 Under his successor Seleucus IV we hear nothing of the east, and with his 
assassination in September 175 B.C. a younger son of Antiochus III assumed the throne 
after having spent more than a decade as a hostage at Rome. 93 One problem in the 
sources with Antiochus IV is either a confusion of events with those of his father, or an 
actual intention on the part of Antiochus IV to follow the footsteps of his father. Just 
as Antiochus III, his son in 166 or 165 invaded Armenia and forced Artaxias, who 
earlier had declared his independence from the Seleucids, to submit and pay tribute. 
Afterwards he went to the Persian Gulf and was active there restoring cities and 
campaigning. He then entered Elymais to plunder a temple, as had his father, but 

in the same way as Euthydemus and other kings of place at this time seems to have started from 

the east, and Antiochus did traverse the Persepolis- Antiochia in Persis. 

Persian Gulf route on his way from Kerman, but 92 On the death of Antiochus HI and the 

this again does not assure us that a Seleucid satrap parallels with that of Antiochus IV, with sources, 

ruled the Persepolis region, which I consider most see Will, supra, Hisloire politique, 2, 200-02. 

unlikely. 9i Comparable to the study on Antiochus III by 

91 For the inscription see Welles, op. cit. [n. 68], Schmitt is the book by O. Morkholm on Antiochus 

24, 32-34, and for a commentary Will, op. cit., 2, IV of Syria (Copenhagen, 1966), esp. 166-80, for 

55. The expedition against Gerrha which took his eastern campaign. 



172 Chapter VI 

perhaps this time in vengeance for his father, but according to the sources he was 
repulsed and left for Media but died of illness on the way at the end of 1 64 B.C. 
The elaborate hypothesis of Tarn that Antiochus IV intended to restore the empire 
of Alexander the Great with a two-pronged attack on the Parthians by himself from 
the west and by his supposed cousin Eucratides from Bactria is amply criticized by 
Morkholm, to which I have nothing to add, except to underline the complete lack of 
evidence for any grand strategy, not to mention Seleucid family relationships. 94 The 
notice of Pliny (VI, 152) about a governor of Mesene called Numenius under King 
Antiochus who won a sea battle ofFthe straits of Hormuz and then a land battle, has 
been placed in the time of Antiochus III or IV, but whichever, it is only an isolated 
incident in the many battles fought by the Seleucids. We next hear of the east with a 
revolt of a certain Timarkhos, who was governor of the 'upper provinces,' at this time 
little more than western Media, against Demetrius the Seleucid ruler from 162-150 
B.C. The coins of Timarkhos are all from Hamadan according to Le Rider, and in the 
prologue to Pompeius Trogus (lib. 34) he is called 'king of Media.' 95 Even though he 
sought alliances with King Artaxias of Armenia and others, he was defeated and killed 
by Demetrius in 160 after extending his rule in Mesopotamia. The much reduced 
'upper provinces' continued under Seleucid rule throughout the reign of Demetrius to 
1 50 B.C. and then under Alexander Balas who had been brought to power by Rome 
and her allies. The coins of Alexander Balas struck at Hamadan did not last more than 
two or three years (150-148 or 7 B.C.), when his coins are followed by those of 
Mithradates I. So the governor-general of the 'upper provinces' Kleomenes, who in 
the Herakles Behistun inscription is called 'saviour,' probably defeated a Parthian 
force in one battle but lost the province shortly thereafter. At the beginning of July 
141 B.C. the Parthians are in Babylonia, according to cuneiform documents. For a 
short period Demetrius II regains not only the plains, but invades Media where he is 
defeated and captured by Mithradates I at the end of 140. All Seleucid possessions in 
the east were lost, but in 130 B.C. Antiochus VII, brother of Demetrius, launched an 
expedition which not only recovered the lowlands with Seleucia and Susa, but he 
may have also established some sort of control in Media. Demetrius was released by 
the Parthians to stir up trouble for his brother Antiochus, but the latter was utterly 
defeated and killed by the Parthians in 129, and this is the complete end of Seleucid 
rule in the east which, however, lasted more than a century and a half. 

THE HERITAGE OF HELLENISM 

The heritage of Hellenism in Iran and Central Asia is difficult to judge, but some 
general remarks can be made. Just as under the Arabs centuries later, the conquerors 
relied on control of the cities on the main trade routes to the east to maintain their 
control, and where there were no cities, they both founded military garrisons. There 
were not enough Greco-Macedonians, or Arabs later, to occupy all cities, but only the 
strategic settlements, beginning with Ecbatana— Epiphaneia— Hamadan and surround- 
ing Media in the west, and continuing to the east with Rhaga-Europos-Rayy, were 

" 4 Tam, op cit , |ch 1, n. 23), 187-91, and c ' 5 Le Rider, op cit. [n. 25), 332, and O. Seel.ed., 

Morkholm, op cit |n. 93), 172-80. Pompeius Trogus, Fragmenla (Leipzig, 1956), 162. 



Alexander the Great and the Seleucids J 73 

held as the key to control of the route to the original 'upper satrapies.' Margiane- 
Antiocheia-Merv was always an important center connecting Khurasan with 
Central Asia and Bactria. Bactria became the center of Hellenism for eastern Iran and 
Central Asia as Media was for western Iran, and in Bactria the Seleucids built the 
largest number of settlements in the east. Along this west-east axis other centers were 
established as Herat and Hekatompylos, but other important settlements were made 
away from this axis in Qandahar, at Alexandria at the foot of the Hindukush north of 
Kabul, and in Sogdiana, but the Media-Bactria line was of paramount importance, 
as the Seleucids did not maintain their rule in Qandahar and probably only 
sporadically in the Hindukush area and Sogdiana and other parts of Central Asia. 
Nonetheless, in the cities and settlements the Greek language, institutions and culture 
reigned supreme. From the excavations of Ay Khanum on the Oxus River, we can see 
a piece of Greece transplanted to Bactria, with no mixture of local influences, and the 
only fusion which took place under the early Seleucids was that of Macedonians and 
Greeks, or other colonists from Asia Minor or the west, such as Thracians, all of 
whom adopted Greek culture. This at first was the ruling class and the natives did not 
participate after the death of Alexander except as they too became Hellenic in culture, 
but even this process took time. 

The Hellenistic world of the Seleucids was one of mercenary armies and loyalty to 
the person of the ruler, who relied on his colonists for support as the latter looked to 
the king for protection. The rule of the Seleucids depended on the loyalty of this 
ruling class, and where it was strong, as in Media and Bactria, allegiance to the house 
of Seleucus developed almost into a religious feeling of colonists towards the ruler as 
their saviour in time of attack. So we find cities giving titles to the Seleucid rulers, 
which has been often described as the Hellenistic ruler cult, and which legitimized the 
absolute power of the ruler, although this dynastic cult in the cities should be 
distinguished from the apotheosis of the dead king by his successor who dedicated 
temples and services to the deified ruler. 96 Natives in cities and settlements surely 
participated in the ruler cult, but hardly the masses in villages or on the land, or 
nomads. Likewise, under the Seleucids Greek culture remained almost the monopoly 
of the colonists and their descendants, and art and architecture were produced by city 
artisans for the ruling class. At the same time the cities were the centers of culture and 
were bound to the rulers, and both provided examples or models for others to follow. 
The 'Hellenization' of the east can be seen best under the Parthians, the heirs of the 
Seleucids, both culturally as well as politically. It was really after the fall of the 
Seleucids that the effects of Hellenism penetrated everywhere, but, of course, then in 
an ever more diluted or syncretic form. 

Since in political organization the Seleucids, on the whole, followed the 
Achaemenids in the east, there was little original Greek contribution in this domain, 
as far as we can see, except for elements of polis organization, which was new to the 
Orient. 97 Unfortunately we have absolutely no information about the numbers of 

96 Every book on Hellenism discusses it; cf. 91 The Greek polis has received much attention; 

Bikerman,op.rif.[n.46],236;F.Taeger,Oi<iram<i, for the Seleucids see esp. Koshelenko, op cit. 

1 (Munich, 1957), 309; Bengtson, Griechische [n. 23], 292 pp., where further references may 

Geschichte [n. 46], 437. be found. 



1 14 Chapter VI 

western colonists in the cities of Iran and the east, and we cannot even guess the 
numbers of natives who took Greek names, learned the Greek language and became 
associated with the conquerors. The statistical analysis of Greek and non-Greek 
personal names in inscriptions from Mesopotamia, the only area in the east where any 
information is available, can be misleading, for a father who had a Greek name by no 
means always named his son also in the Greek manner. Unlike the acceptance of 
Islam, many centuries later, which did involve the transfer from one religious 
community to another, and names can tell us about conversions, the acceptance of a 
Greek name implied no conversion of any kind, and the evidence only shows that 
Hellenism was generally accepted by the urban population side by side with native 
cultures, and only then did mixtures occur which eventually produced a kind of 
syncretism before Hellenic elements became absorbed. Obviously there was a strong 
element of prestige and privilege involved in the acceptance of Hellenism by a native, 
not the least of which was the chance to participate in the ruling circles of a city or 
even at the court of the Seleucid king. All the time we must remember that our 
evidence comes from Mesopotamia, with only a rare inscription from the Iranian 
plateau. ( 

In religion one must be even more careful in drawing parallels from the western 
part of the Seleucid Empire for Iran and the east, since we have no information about 
the Zoroastrian religion under the Seleucids. The Greeks brought with them not only 
visible institutions such as the gymnasium and theater in their cities, but also in this 
period undoubtedly social and religious clubs, societies, and other organizations. In 
the beginning we can imagine that there was little if any contact between the Greek 
priests and their Iranian counterparts, but in time the Greek propensity for 
identifying local deities and local stories with their own, probably influenced Iran as it 
did other areas and religions in the Near East. Eddy has suggested that the break-up of 
the Achaemenid Empire in Iran meant the rise of local cults and practices in 
Zoroastrianism, with varying degrees of religious opposition to the conquerors. 98 
Both the religious and cultural unity, such as it was under the Achaemenids, was 
broken under the Seleucids and the Parthians, not to be restored, and then 
incompletely, until the Sasanians. It seems as though the Seleucid control of the route 
from Media to Bactria divided the Iranian area between north and south, such that 
outlying centers developed their own traditions, their own variants of the Aramaic 
script for their native languages, and some differences in Zoroastrianism even though 
we cannot recover them. Under the Seleucids, the Sogdians, Khwarazmians and 
Bactrians in the east developed independently, as did the Armenians, Georgians and 
Albanians in the Caucasus, as well as the isolated Caspian peoples. Presumably, in 
Persis the traditions of the Achaemenids, in religion as well as in society, were best 
maintained. Eddy (p. 330) maintains that local resistance to the rule of foreigners was 
based on the desire to throw them out, to end social and economic exploitation and to 

98 Eddy, supra. King is Dead, 81 , 328. His use of other priestly writings but apocalypses and sibyls 

the Bahman Yasht and the Oracle of Hystaspes as are notoriously difficult to pin down to actual 

sources for a widespread religiously organized events, other than long-standing resentments or 

opposition to the Seleucids, in Persis, is unconvinc- the like. Cults and temples proliferated in Iran 

ing. This is not to deny the reality of an anti- under the Seleucids, and one must not use later 

Alexander-Seleucid-Parthian bias in these and texts as accurate indices for the Seleucid period. 



Alexander the Great and the Seleucids 175 

regain power to protect native law and religion. The last, however, was hardly a 
compelling motive in Iran except for the natural desire to be dominant, but any 
strong impetus to recreate a central state to replace the Seleucids was absent. Perhaps 
just as Alexander had brought an end to the heyday of the independent Greek polis, his 
death also really brought an end to the idea of a unified, centralized government in the 
east, the union of religion and state, although the reality of this belief under the 
Achaemenids is also open to question. In any case, we find no more a unified 
Zoroastrian opposition to Seleucid rule than a unified striving to re-establish an 
Iranian Empire." This was not to come for many centuries in a new world of 
organized state religions. 

The activities of Greek savants such as Berossos in Mesopotamia cannot be 
discussed here, but in Iran we have no comparable names. It is not our purpose to 
examine Iranian influences on the Greeks, but they too were impressive. Just the 
influx of new foodstuffs such as the peach 'the Persian apple,' citron 'Medic apple' and 
alfalfa 'Medic grass,' to name a few, indicates the interchange between peoples in this 
period. 100 The great expansion of trade with the east under the Seleucids must have 
impressed the natives and influenced them. But this, as in the case of the Fine Arts, is 
most probably a post-Seleucid development, since the century and more of Seleucid 
rule on the Iranian plateau is a time of separation, Greek art and Achaemenid art in 
western Iran and Greek art with local, folk art in Bactria and elsewhere in the east. 
The effects of Hellenism which led to a Greco-Iranian syncretic art are strong after the 
Seleucids had left the east, but they left behind cities and local dynasts, some of whom 
perhaps were even more pro-Hellenic in their sentiments than were the Seleucids. In 
the arts, this Greco-Iranian synthesis about the beginning of the first Christian 
millennium developed into Parthian art in the west and Greco-Buddhist or 
Gandharan art in the east, but that is a later story. 

99 The idea that under the Achaemenids church unlikely. The opposition of Alexander to the 

and state were highly organized and centralized is, exposure of corpses in Bactria (Strabo XI, 517) 

of course, absurd, and the Sasanian belief that the cannot point to a general opposition to Zoroas- 

period from Alexander to Ardashir was a period of trianism but merely to a custom repugnant to the 

the rejection of that ideal is also false. Zoroastrian- Greeks, but even this story is suspect, 
ism had a long period of steady development to l0 ° See Forbes, op. cit., 2 [ch. 3, n. 8], 87; 4, 46- 

the Sasanian period and Greek religious ideas and 55, and elsewhere. Some Iranian philosophers or 

rituals, on the whole, were not very different from scientists existed, but we have no information 

the Iranian, so the idea of an organized Iranian about them, 
religious opposition to the Greeks is highly 



CHAPTER VII 
GRECO-BACTRIANS, SAKAS AND PARTHIANS 



Literature: Over thirty years ago writings on the Greco-Bactrians were based only on numismatics, as well 
as a few art objects, mostly silver vessels, but it was the remarkable series of coins which first attracted the 
attention of scholars. The history of research in previous publications was summarized by K. V. Trever in 
her Pamyatniki Greko-Baktriiskogo Iskusstva (Moscow-Leningrad, 1940). On a much wider scale, and 
approaching the Greco-Bactrians from the side of India, were the two volumes of A. Foucher, La Vielle 
Route de I'Inde de Bactres a Taxila, MDAFA, 1 (Paris, 1 940-47), 2 vols. This pioneer work is a survey of the 
geography and history of ancient Afghanistan and northwest India until the Islamic expansion into India 
under the Ghazna vids at the beginning of the second millennium of our era. Here he described in detail his 
position on the origin of 'Gandharan art,' which he called Greco-Indian art, as the residue of Hellenistic 
influence on Buddhist northwest India which produced the flowering of Buddhist art known as 
'Gandharan' in the first century of our era. This assertion was opposed by other scholars who believed that 
Gandharan art had no local predecessors but was 'planted in a vacuum and influenced by provincial Roman 
art, according to the summary of M. Wheeler, "Gandhara Art: a note on the present position," in the 8th 
Congres international d'archiologie dassique of Paris 1963, published as La rayonnement des civilisations 
grecque et romaine sur les cultures ptriphiriques (Paris, 1965), 560. The links between the Greco-Bactrians 
and Gandharan art even after the excavations at Ay Khanum still present problems, and controversy over 
the western or Roman influences on Gandharan art has npt been stilled. Most likely there continued an 
echo of Greco-Bactrian art which contributed to a new art in the service of Buddhism sponsored by the 
Kushan rulers, who had contacts with the Roman Empire, which influence contributed much to the 
formation of Gandharan art. This will be elucidated in the chapter on the Kushans. 

Parallel to the controversies of the art historians went research in numismatics, for the chronology and 
history of the Greco-Bactrians was based almost entirely on coins, dated by style and the symbols and 
monograms on them. Many new coin types have been found in recent years and the order of rulers has 
been rectified several times. Since World War II the work of numismatists and art historians has been 
supplemented by archaeological excavations in present Soviet Central Asia, Afghanistan and the 
subcontinent of India which have given new vistas especially on the material culture of the area where the 
Greco-Bactrians ruled. The most important of these excavations has been the Hellenistic city of Ay 
Khanum, identified as Alexandria on the Oxus by the excavator Paul Bernard in R. Andrim et P. 
Bernard, Tresor de monnaies Indiennes et Indo-Grecques d'Ai' Khanoum," RN, 5 serie 15 (1973), 238- 
89, and 16 (1974), 7-41, or as Eucratideia. I. T. Kruglikova, however, in "Novye antichnye pamyatniki 
Yuzhnoi Baktrii," in I. R. Pichikyan, ed., Antichnost i antichnye traditsii v kulture i iskusstve narodov 
Sovetskogo Vostoka (Moscow, 1978), 270, identifies Dilberdzhin Tepe northwest of Balkh as Eucratideia. A 
new site on the Oxus, Takht-e Sangin, promises new Greek remains; cf. B. A. Litvinskiy and I. R. 
Pichikiyan, The Temple of the Oxus," JRAS (1981), 133-67. The results of excavations at Ay Khanum 
since 1964 have been published almost yearly in CRAI, and sometimes twice a year as in 1975, pp. 167- 
96 and 287-322, as well 3sJA and elsewhere. The results of the excavations 1965-68 are found in P. 
Bernard, Fouilles d'Ai Khanoum, MDAFA, 21 (Paris, 1 973) ; cf. C. Y. Petiot-Biehler et P. Bernard, Tresor 
de monnaies grecques et greco-bactriennes trouve a Ai' Khanoum," RN 6 serie, 17 (1975), 23-70. Other 
sites have had Greco-Bactrian strata such as Dalverzin Tepe on the Surkhan Darya in southern 
Uzbekistan, where the finds of wall paintings and sculptures, however, are from a later period and the 
Greco-Bactrian remains are limited to a few stone column bases and pottery. Cf. G. Pugachenkova, Les 
Trisors de Dalverzine Tipi (Leningrad, 1978), and in Russian, her Dalverzintepe, kushanskii gorod na yuge 
Uzbekistan (Tashkent, 1978), 240 pp. A survey of recent Soviet archaeological work in Central Asia is 
given by B. A. Litvinskii, "Problemy Istorii i Istorii Kultury Drevnei Srednei Azii 1967-77," VDI, 4 
(1 977), 73-92. A veritable diarrhoea of articles and books on the art and archaeology of the Kushan period, 
as well as materials on earlier periods, has appeared in the Soviet Union, but it is impossible to even find 
many of the publications. In Taxila, Pakistan, the town called Sirkap apparently was a foundation 
of the Greeks from Bactria. See J. Marshall, Taxila, 1 (Cambridge, 1951), 112. A survey of the many 
publications over a time period of just over two years on the Indo-Greeks and the Kushans is given by G. 



178 Chapter VII 

Fussman, "Chronique des etudes kouchanes (1 975-77)," J A (1 978), 41 9-36. The section of 'Asie Centrale' 
in the Abstracta Iranica, supplements to SI, gives summaries of the publications in this area; see also the 
section on Achaemenids and Greeks in The Archaeology of Afghanistan, ed. by F. R. Allchin and N. 
Hammond (London, 1978), 187-233. The bibliographies in that publication as well as in Numismatic 
Literature of the ANS are very useful. 

The three basic general works on the Greeks are W. W. Tarn, The Creeks in Bactria and India, 2nd ed. 
(Cambridge, 1951). A. K. Narain, The Indo-Greeks (Oxford, 1957), with Hellenistic and Indian 
viewpoints resp. of the authors coloring their surveys, and F. Altheim and R. Stiehl, Geschichte 
Mittelasiens im Altertum (Berlin, 1970), with many stimulating idiosyncracies; cf. the long review by R. 
Schmitt in WZKM, 67 (1975), 31-91. All of the general works must be read with caution, for the 
surmises in them cannot be considered as facts, while the varied approaches of the authors must be 
remembered. Tarn was a Classicist who reads perhaps too much into reconstructed variant readings in 
Greek and Latin texts, while Narain relied overly on numismatics, and Altheim followed Tarn but wove 
personal notions into his narrative. For a political history of the Greco-Bactrians, numismatics is certainly 
more important than elsewhere, and many publications on coins must be consulted. A good survey is 
found in A. N. Lahiri, Corpus of lndo-Creek Coins (Calcutta, 1965), to be supplemented with the 
compilation of M. Mitchiner, Indo-Greek and Indo-Scythian Coinage (London, 1976), which, however, 
contains some questionable specimens as well as conclusions. Catalogues of coins are also important, 
especially those with large collections such as the British Museum by Percy Gardner (London, 1886), the 
Lahore Museum by P. B. Whitehead (Oxford, 1914), the Indian Museum in Calcutta by Smith (Oxford, 
1906) and others, but hoards are of most importance, since their provenance is known. For the Greco-/ 
Bactrian period the Oxus treasure hoard is the largest and earliest, and a bibliography of it is given in M. 
Thompson, O. Morkholm and C. M. Kraay, An Inventory of Creek Coin Hoards, ANS (N.Y., 1 973), 263^ 
64. The Ay Khanum hoard has been mentioned above, and to the hoards listed by Thompson, et al., the 
following should be noted: Mir Zakah northeast of Gardiz; cf. R. Curiel et D. Schlumberger, Trhors 
monhaires d' Afghanistan, MDAFA, 14 (1953), 65-91, Qandahar by D. MacDowall in Afghan Studies, 1 
(London, 1978), 51 , Butkara in Swat, Pakistan by R. G6bl, in A Catalogue of Coins from Butkara, 1 (Rome, 
1976), as well as minor finds in many Soviet and Indian excavations. 

The period of nomadic invasions, primarily of the Sakas (c. 130 B.C.-A.D. 50) has produced perhaps a 
smaller literature on coinage and archaeology than the Greeks. For numismatics, see the compilation by 
Mitchener noted above as well as the catalogues of collections already mentioned, plus A. K. Sri vastava, 
Catalogue of Saka-Pahlava Coins of North India (Lucknow State Museum, 1 972). See also G. K.Jenkins and 
A. K. Narain, The Coin-types of the Saka-Pahlava Kings of India (Numis. Soc. of India, Varanasi, 1957) and 
many articles in the fournal of the Numismatic Society of India. The doctoral thesis of K. Walton Dobbins 
on the coinage and epigraphy of the Sakas and Pahlavas at the Australian National University at Canberra 
in 1972 will be published in Islamabad, Pakistan, and it will bring up to date work on this period. On the 
migration of the Sakas to Seistan, all material has been assembled by P. Daflina, L'Immigrazione dei Sake) 
nella Drangiana (Rome, 1 967), while an analysis of nomadic movements from their burials is found in 
A. M. Mandelshtam, Kochevniki na puti v Indiyu (Moscow, 1966), although his main excavated site of 
Tulkhar in the Bishkent valley of Tajikistan dates from the Kushan period. The wealth and far-flung 
connections of the nomad chiefs may be seen in the 20,000 gold pieces from graves of early Kushan 
leaders. See V. 1. Sarianidi, "The Treasures of Golden Hill," A] A, 84 (1 980), 1 25-31 . There have been no 
large exclusively Saka sites excavated, although strata from the nomadic period have been found in 
excavations in Soviet Central Asia and on the subcontinent. 



THE GREEKS IN CENTRAL ASIA AND AFGHANISTAN 

The revolt of Diodotus, governor of Bactria under the Seleucid Antiochus II, cannot 
be dated, although much has been written about the circumstances of the break away 
of the eastern part of the Seleucid Empire. We have seen that the coinage of Antiochus 
II at Bactra changes to his portrait with the legend 'King Diodotus' and finally to the 
issues of Diodotus with his own portrait and legend, which implies that the process 
leading to independence was not a sudden break. The fraternal war between Seleucus 
II and Antiochus Hierax, after the death of Antiochus II in 246 B.C., probably provided 
the occasion for a definite and complete break, since the era of the breakaway of the 



Greco-Bactricms, Sakas and Parthians 1 79 

Parthians, who were contemporaries of Diodotus, began from 247 B.C. less than a year 
before the death of Antiochus II. Wolski has written many articles on the rise of the 
Parthians which he has insisted took place in 238 B.C., for him a year after the revolt of 
Diodotus. 1 His arguments from the text of Justin (XLI, 4) that the revolts of both 
Parthians and Bactrians took place during the struggle between Seleucus II and 
Antiochus Hierax have convinced all Classicists, but the absence of coins of Seleucus II 
in the east, and the date of the beginning of the Parthian era remain to be explained. A 
prolonged period of separation from Seleucid rule with a gradual break-away seems 
the best answer to this problem in spite of the arguments of Wolski. 

When we remember that both Alexander and Seleucus took Bactrian or Central 
Asian wives, the possibility of an early and continuing feeling of accommodation 
between the natives and the conquerors is not an unlikely surmise. Whether the 
policy of the Greco-Bactrians was one of sharing of rule and the assimilation of 
conquerors and natives, as opposed to a Seleucid policy of pure colonialism with 
Greek supremacy in every domain, as suggested by Tarn and others, is impossible to 
substantiate. Nonetheless Greek influence was not only stronger but more lasting in 
Bactria than elsewhere in the east, but the Greco-Bactrian kingdom must be 
considered one of the great Hellenistic kingdoms, together with the Seleucids and 
Ptolemies. 2 

Diodotus, according to Justin, died a short time after becoming king and was 
succeeded by his son of the same name, who made peace with the Parthian king to the 
west. According to Narain {supra, The Indo-Greeks, 16) the portraits of the elder and 
the younger Diodotus appear on different coins and both of them issued "coins with 
the name of Antiochus," followed by "money with their own name, type, and portrait 
complete." Why should Diodotus II, however, strike coins with the name of 
Antiochus on them after his father had declared his independence and had proclaimed 
himself king? It is hardly conceivable, as Narain proposes, that both father and son 
played with the coinage, sometimes putting the name of Diodotus on a portrait coin 
of Antiochus II and sometimes having a portrait of either Diodotus on coins, with the 
name of Antiochus. More likely, as suggested above, was a changed relationship with 
the Seleucids, which is difficult to reconstruct solely on coin types. It has been 
suggested that father and son participated in joint rule in the period when Bactria was 
still subject to Antiochus II, and both issued coins in the name of the Seleucid ruler, 
and later both changed to their own portraits and common name. Other instances of 

' J. Wolski, "Les Iraniens et le royaume greco- material remains, not to mention the use of 

bactrien," Klio, 38 (1960), 1 17, "Der Zusammen- Aramaic later rather than Greek as the alphabet of 

bruchderSeleukidenherrschaftimIranim3.Jh. v. writing. Cf. articles by Staviskii (esp. p. 211), 

Chr." in Der Hellenismus in Mittelasien, ed. by F. Lelekov (esp. p. 228) and Vorobeva (234-35) in 

Altheim (Darmstadt, 1969), 188-254, and with a Pichikyan, supra, Antichnost. To deny the Bactrian 

detailed analysis of the text of Justin (Trogus) in Greeks the epithet of a 'Hellenistic monarchy' on 

"Untersuchungen zur frUhen parthischen Ge- the grounds that they acted differently from 

schichte," Klio, 58 (1976), 439-57. Justin has eodem Ptolemies and Seleucids, as Narain, supra, Indo- 

lempore (of the fraternal war) eliam Diodotus, milk Greeks, 11, argues, ignores the evidence of Ay 

urbium Baclrianarum praefectus, defecil regemque se Khanum and a certain parallel with the Bosphoran 

appellari iussil. kingdom of south Russia. The Greco-Bactrians 

2 Bactria was surely the center of Greek were absorbed in India, but they were the heirs of 

colonization in the east, for Sogdiana and Khwar- Alexander as were other Hellenistic kingdoms, 
azm show far less Hellenistic influence in their 



180 Chapter VII 

joint rule and joint coinage exist but here we have no proof. Unfortunately, all 
argumentation about the rise of the Greco-Bactrians and the Parthians as well is 
subjective. For example, J. Wolski in many articles has argued that it is unthinkable 
that Antiochus II would have allowed a break-away of the east from Seleucid control, 
and it must have been in the time of the fraternal war of Seleucus II that both Bactria 
and Parthia declared their independence. 3 His dating of the events as 239 B.C. for the 
revolt and assumption of kingship by Diodotus, with the Parthians a year or more 
later, may by correct as the final break, but de facto the independence of Bactria must 
have been a reality more than a decade earlier. Wolski's further suggestion that the 
Greco-Bactrian revolt was one of Greeks and Iranians against Macedonians finds no 
support in the sources and is difficult to follow, at least from the position of the native 
Iranian population who hardly distinguished between the two foreign groups. Other 
scholars have argued that only one Diodotus existed, but the statement of Justin as 
well as the coins with old and young portraits suggest that two rulers with the same 
name are needed to fill the gap in time between the middle of the century and the 
time of Euthydemus and Antiochus III, rather than one king represented in youth and 
old age. On the other hand, Justin's dating of the break-away of the east under 
Seleucus II, in the first Punic war when Lucius Manlius Vulso and Marcus Attihtis 
Regulus were consuls of Rome, cannot be considered reliable as Narain (p. 14) 
pointed out, since the consuls were in office in 256 B.C., while Seleucus only began to 
rule in 246. All attempts to change the names of the consuls or to explain the 
discrepancy in other ways only shows that approximate dates were recorded, and the 
sources are not concerned with precisely dated events. 

The Greco-Bactrians probably ruled Sogdiana, and probably also the oasis of Merv 
according to Strabo (XI, 517), but the existence of some coins of Diodotus 
(presumably the father) with the appellative soter 'savior' hardly can be interpreted as 
a reference to his conquests, although the defeat of a nomadic invasion from the north 
cannot be excluded. Justin's statement that the first Parthian ruler feared Diodotus I 
but made peace with his son implies an aggressive policy of the first Greco-Bactrian 
king, but beyond this surmise one cannot go. The theory of Tarn (op. cit. supra, [ch. 1 , 
n. 23], 74) that Diodotus II was married to a Seleucid princess has no basis in the 
sources. Sometime, perhaps about 230 B.C., a certain Euthydemus from Magnesia 
(which Magnesia is unknown) apparently overthrew Diodotus II, for Polybius (XI, 
34) says he destroyed the descendants of those who had revolted (against the 
Seleucids). Narain is probably correct in rejecting Tarn's hypothesis that the Seleucid 
princess, widow of Diodotus I but not the mother of Diodotus II, married one of her 
daughters to Euthydemus and incited him to overthrow Diodotus II on behalf of the 
Seleucids. In any case, no evidence exists for this elaborate reconstruction, and the 
next bit of information comes from Polybius (X, 49) about the invasion of the 
Seleucid king Antiochus III. The latter defeated Euthydemus on the Hari Rod, which 
at least indicates that the Greco-Bactrian kingdom extended to Herat. Afterwards 

Wolski, "Untersuchungen," [n. 1 ], 444, where revolts under the rule of Antiochus II. To change 

further bibliography is given. Wolski bases his Strabo's (XI, 515) Euthydemus to Diodotus and 

chronology on a statement of Strabo (XI, 515) that then use Strabo as a source for the date of the 

the revolt of Diodotus took place before that of the Greco-Bactrian revolt, as Wolski and others do, is 

Parthians, against Arnan's Parlhika, who puts both not convincing. 



Greco-Bactrians, Sakas and Parthians 181 

Antiochus pursued Euthydemus to his capital, called Zariaspa, and besieged him there 
for two years. The explanation of the name of the city as 'having golden horses,' 
which was taken from an appellation of Bactra relating to the goddess Anahita, 
proposed by Altheim, may be correct, but surely the main city of Bactria is meant, 
and we do not have two different towns. 4 According to Polybius, Euthydemus made 
peace with Antiochus and retained his title of king, invoking the fear that nomads 
would destroy them if they did not make peace. The two rulers concluded an alliance 
and Antiochus promised one of his daughters in marriage to Demetrius, son of 
Euthydemus. Having obtained some elephants, he crossed the Hindukush mountains 
and made or renewed an alliance with Sophagasenus, an Indian ruler, after which he 
returned to the west in 206-205 B.C. Afterwards no Seleucid ruler even came near to 
the Bactrian Greeks, for they were separated by the Parthians. 

The find spots of coins may aid in establishing geographical limits of rule, but 
primarily hoard finds of copper coins, which do not travel, are reliable in this matter. 
In Soviet Central Asia coins, or copies of coins, of later Greco-Bactrians, such as 
Heliocles, have been found, but coins of Diodotus and Euthydemus are conspicuous 
by their absence. This can mean either that such coins had a small circulation or they 
were not in current use north of the Oxus River. Copper coins with the name 
Diodotus on them have been found in Afghanistan but they are not plentiful, so in 
any event we may conjecture that the coins had a small circulation. Whether 
Sogdiana, including the Ferghana valley, was under the rule of Diodotus cannot be 
determined, but at some time Greco-Bactrian rule in some way included this 
northern province. 5 Beyond Sogdiana there is little evidence for Greek rule and the 
finds of imitations of coins of Heliocles in eastern Ferghana tell us little about earlier 
Greco-Bactrian rule there. 6 The mere finds of coins, of course, must not be considered 
evidence of rule, solely from the find spots or the presence of certain types; for 
example, the uncovering of a coin of Diodotus and four of Eucratides from an 
excavation in the Caucasus region cannot attest to Greco-Bactrian rule there. 7 
Consequently the use of numismatics in reconstructing history must be regarded with 
great circumspection. Since coins cannot aid us in reconstructing the northern 
borders of the Greco-Bactrian state under the early kings, we must turn to India and 
the conquests of Demetrius, son of Euthydemus. 

Strabo (XI, 516) quotes Apollodorus of Artemita, who says the Greeks of Bactria 
became so powerful because of the fertility of their land that they conquered Ariane 

4 Altheim and Stiehl, Geschichte Mittelasiens [ch. Tajikistan were copper, but probably later copies 
6, n. 49], 403. The statement of Pliny (VI, 48) that of Eucratides' coppers. 

the name of the river was Zariasta (sic for Zariaspa) 6 Cf. V. M. Masson, "Drevnebaktriiskie mon- 

may be true, while Ptolemy's (VI, 11,6) designa- ety, chekanennye po tipu tetradrakhm Geliokla," 

tionofa people called Zariaspi living there tells us EV, 11 (1957), 63-75. Imitations of coins of 

little other than that they were Iranians. Heliocles probably were struck by the nomads 

5 For places in Afghanistan where coins of who ended Greco-Bactrian rule, and Masson 
Diodotus are preserved see Lahiri, supra. Corpus suggests this find indicates an opening of the 'silk 
under 'Diodotus.' Coins of Heliocles and Eucra- route' to China not by the Greeks but by nomads, 
tides have been found north of the Oxus in several 7 S. A. Dadasheva, "Greko-Baktriiskie monety, 
sites. For a bibliography of the sites with such coins naidennye na territorii Kavkazskoi Albanii, in 
see B. Ya. Staviskii, Kushanskaya Baktriya (Mos- Baktriskie Drevnosti, ed. by V. M. Masson (Lenin- 
cow, 1977), 239^*1. The coins of Eucratides from grad, 1976), 106-09. 

the Tulkhar kurgan in the Bishkent valley of 



182 Chapter VII 

and India, and some tribes were subdued by Menander while others by Demetrius 
son of Euthydemus. Until the reign of Demetrius most scholars agree on the general 
outlines of the rule of the Greek kings of Bactria, but with Demetrius, and more with 
his successors, numismatists dispute the sequence of events. Tarn sought to extend the 
conquests of Euthydemus into Chinese Turkestan (pp. 84—87) and those of his son 
deep into India (pp. 142-45), but all the evidence adduced for these far-flung 
conquests, such as coins and seals found in the two areas, possible Greek words in the 
documents in Prakrit from Sinkiang, and the import of nickel from China, are all 
quite weak evidence for Greek rule so far east. It is true that Apollodorus of Artemita 
in Strabo (XI, 516) says that they extended their rule as far as the Seres and Phruni, but 
the identity of the latter has been disputed. In any case this statement implies that their 
rule extended only to Chinese Turkestan, which could have really meant the 
Ferghana valley or the Alai valley, or even the Pamirs. 8 Whichever direction the 
Greco-Bactrians expanded, their hold over outlying territories to the north and east 
of their center in Bactria must have been tenuous. Since only the coins of later Greco- 
Bactrian kings are found in territories north of the Oxus, an expansion later than 
Euthydemus might be suggested. India, however, presents other problems, fqf 
Narain, while conceding that Demetrius conquered Arachosia and possibly part d( 
Seistan-Drangiane, does not include the mountainous part of present Afghanistan. 
This has been convincingly refuted by another numismatist, A. Simonetta, who 
showed that the arguments of Narain for the occupation of the Paropamisos only later 
by a Demetrius II are unacceptable. 9 We may assume that Demetrius I did occupy the 
Paropamisos and Arachosia, the latter primarily because of a notice in the Parthian 
Mansions of Isidore of Charax (paragraph 19) that such a city existed there, probably 
named after Demetrius I. But Narain (p. 43) is probably right in rejecting extensive 
conquests of Demetrius I in India against Tarn and Altheim. Demetrius probably did 
not rule later than c. 185 B.C., but the many successors present problems. 

The activities of counterfeiters or forgers, both ancient and modern, have greatly 
complicated the studies of numismatists, who depend on coin styles, monograms and 
variations in titles to reconstruct the history of the later Greco-Bactrians, but even 



8 Cf. Tarn, op. cil [ch. 1, n. 23], 84, who refutes 9 Cf. A. M. Simonetta, "A new essay on the 

the identification of the Phruni or Phuni as Huns, Indo-Greeks, the Sakas and 'the Pahlavas," EW, 9 

while Narain, 170-71, identifies the Seres, to be (1958), 157. Narain, supra, lndo-Greeks, 30-31, 

read Sures, as Chinese Su-le or Kashgar and the bases his claim on the absence of coins of Demetrius 

Phruni as people inhabiting the valley of Tash- 1 from the Kabul region and in India, but coins of 

kurghan. These identifications were refuted by J. Demetrius 1 are rare in general and we simply have 

Harmatta in "Sino-Indica," AAH, 12 (1964), 10- no coin finds from excavations near Kabul, and the 

11, but his further assumption that the Greco- hoard from Mir Zakah near Gardiz is too far to the 

Bactrians ruled in Sinkiang is only a guess. His east. Demetrius I may not have reigned long and 

reconstructed readings of inscriptions are not we must be careful in assuming that wherever he 

convincing. Altheim's discussion (op. cil. [n. 1 ], ch. went he at once introduced a new coinage. The 

20) of the two peoples is an exercise in philological curious elephant scalp headdress on his coins does 

ingenuity but unconvincing historically. J. Har- imply southern conquests and, since the com- 

matta, "Sino-Indica," AAH, 12 (1964), 12-13, memorative coins of Agathocles dedicated to 

believes the Greco-Bactrians under Demetrius did Demetrius call him Aniketos, this implies a series of 

control Sinkiang and the 'silk route' to the east, but victories somewhere, 
his reasons here too are unconvincing. 



Greco-Bactrians, Sakas and Parthians 183 

without forgeries there are problems. 10 The successors of Demetrius I may have been 
related to him, or there may have been a second Demetrius if one judged by coins. 
The numismatic evidence is not unequivocal, but in general those coins struck on the 
Attic standard of weight only using Greek legends come from Bactria, while the coins 
on the Indian (or lighter) standard with bilingual legends in Greek and Kharoshthi 
were struck south of the Hindukush. 1 ' Coins of both types exist for a Demetrius with 
different styles, and the question is whether we have a Demetrius I and another II. The 
matter is complicated by the statement of Justin (XLI, 6) that Eucratides, a successor of 
Demetrius I "was besieged by Demetrius, king of the Indians with an overwhelming 
force, but Eucratides escaped after a five-month siege and subsequently conquered 
India." The question naturally is which Demetrius, if there be two? Did Eucratides 
revolt against Demetrius I and his two commanders Menander and Apollodotus, as 
Tarn thinks, or against Demetrius II as Narain concluded? There is not agreement on 
the sequence of rulers following Demetrius I, and one must rely on comparisons of 
monograms, facial features and other items on the coins to make conjectures. One 
reasonable guess is the following: since Demetrius I was a young man when 
Antiochus III invaded Bactria in 206 B.C., and Eucratides was the one who overthrew 
the family of the Euthydemids about the time that Mithradates I of Parthia came to 
power about 170 B.C., according to Justin (XLI, 6), and if it were Demetrius I whom 
Eucratides overthrew, the coins would necessarily portray a man at least in his fifties, 
and the coins of the first Demetrius do show a man of such an age. Eucratides, 
however, if we are to believe the evidence of the widespread finds of his coins, plus the 
statement of Justin (XLI, 6) that Mithradates I of Parthia conquered Media about the 
time of Eucratides' death, lived until about 148 B.C., when the Median conquest took 
place, and thus he had a reign of over twenty years. 12 In any case, whether we have 
two rulers with the name Demetrius or only one, in the period c. 170-150 B.C. we 
may postulate that a split in the Greco-Bactrian domains occurred. It seems that at 
least two families now dominated the scene, that of Eucratides in the north and the 
successors of Euthydemus and Demetrius to the south of the Hindukush. One of the 
latter family may have been Antimachus Theos, who possibly ruled north of the 
Hindukush when Demetrius I or II was busy to the south. He was the first king to call 

10 For a discussion of such forgeries see Lahiri, Caucasem, another at Pushkalavati (the present 
supra. Corpus, 62-68. See also G. K. Jenkins, "A plain of Peshawar) and another at Taxila, while 
Group ofBactrian Forgeries," RN 6 serie, 7 (1965), other mints were rare or temporary, which 
51-57. proposal makes sense. 

11 A D H Bi var in a series of articles devoted to ' 2 The date of the Parthian conquest of Media 
the monograms on the coins, on the other hand, should be after 148 B.C. according to the dated 
has shown that coins solely with Greek legends and Greek inscription on the bas-relief of Herakles at 
based on Attic weight could have been struck Behistun, mentioning a satrap of the 'upper 
south of the Hindukush to pay troops, or for provinces' called Cleomenes. Everyone had as- 
prestige value. Cf. his "Indo-Bactrian Problems," sumed that this satrap could only have been a 
NC V, seventh series (1965), esp. 104-05, and his Seleucid officer, but it is conceivable that he may 
The Sequence of Menander's drachmae," JRAS have continued to hold this title under Parthian 
(1970), 128-29. His conclusion that the mono- rule until Mithradates appointed Bacasis in his 
grams either represented mints or mint masters, place, ace. to Justin (XLI, 6). Another possibility, of 
more or less in the same spatial locale, is confirmed course, is a short recrudescence of Seleucid rule in 
by his studies of the monograms. South of the Media, after an initial conquest by Mithradates I, 
Hindukush one main mint was at Alexandria ad but more data is needed to decide. 



184 Chapter VII 

himself Theos 'god,' and the first to issue commemorative coins with a bust of 
Diodotus UtoTTjpos on the obverse and the legend BAEIAEYONTOZ 0EOY 
A NTIMAXOY on the reverse, or with a bust of Euthydemus with the same legends. 
Numismatists have conjectured that Antimachus was a younger brother of 
Demetrius I or II, but at least a member of the same family. There also may have been 
a Euthydemus II as well as a Demetrius II because of varied coin types, but the 
proliferation of names on coins now makes the reconstruction of a sequence of rulers 
of one or the other family very difficult. 

Eucratides, whose origin is unknown, may have been a governor of one of the 
northern or western provinces of the Greco-Bactrian state, such as Sogdiana or Herat 
before his revolt. After attaining power he struck a commemorative coin probably 
showing his parents, a certain Heliocles and Laodice alone wearing a diadem, 
indicating that she may have been a princess of the Diodotus or Euthydemid line. 
Eucratides was successful in expanding his power to the south of the Hindukush, but 
he lost two provinces in the west, Touriva and Aspiones, to the Parthians, according 
to Strabo (XI, 517). Tarn (op. cit. supra, [ch. 1, n. 23], 88) accepted a proposed 
correction of Strabo's text to read Tapuria and Traxiane, about present day Borujjrd 
and Meshhed respectively. Tarn's correction is hardly acceptable, however, but there 
is no way to identify these two 'satrapies' except to note that the size of the satrapy had 
declined, perhaps even more than Tarn thought, and the two most likely would be in 
the Merv-Herat areas rather than farther west close to the Parthian homeland. ' 3 Tarn 
claimed that Eucratides was a cousin of the Seleucid Antiochus IV, creating a 
fascinating story of the attempt of Antiochus IV to revive the empire of Alexander 
the Great, and he began his dream by sending his general Eucratides, governor of the 
'upper provinces,' who overthrew Demetrius (op.cit. supra, [ch. 1, n. 23], 195). All of 
this reconstruction starts from a Greek inscription from Babylon naming Antiochus 
IV the 'savior of Asia,' and Tarn then weaves a tale of Eucratides marching from 
Babylon eastward through Seistan to Bactria, on the way defeating sub-kings of 
Demetrius, Agathocles in Seistan and Antimachus in Herat. Unfortunately Tarn's 
vivid imagination is not based on sources either literary or numismatic. Likewise the 
'so-called' dates on coins of Plato and Heliocles have been identified as monograms, 
and Altheim's surmises (supra, Mittelasien, 57-63), based on these dates, cannot be 
accepted. We do have the name of the city Eucratideia in Bactria from Ptolemy and 
Strabo, evidence of rule here, and further Strabo (XIV, 686) says that Eucratides ruled 
a thousand towns in India, a standard number but nonetheless indicative of his 
conquests there. Justin (XLI, 6) says the Bactrians became fatigued fighting the 
Sogdians, the people of Seistan, Herat and India, which may in this case mean the 
conquests of Eucratides against such rulers as Antimachus, Pantaleon and Agathocles, 
one or all of whom may have ruled before or concurrently with Eucratides. Since 
coins of Euthydemus, Antimachus and Eucratides have been found in greater 

1 Tarn corrected the text of Strabo even more to have read 'the Tapuria of Aspiones (and 

in his "Seleucid Parthian Studies," Proceedings of the Traxiane).' This is ingenious, especially the identi- 

British Academy, 16 (London, 1931), 22-24, and ficationof Aspiones as a personal name, but the rest 

identifies the two words as reference to one is unconvincing. Likewise Altheim's (op. cit. [n. 1], 

satrapy, eastern Tapuria or Astauene. He considers 577) identification of the first name as Turan is 

the original text of Apollodorus, copied by Strabo, unacceptable. 



Greco-Bactrians, Sakas and Parthians 185 

quantity than other Bactrian rulers north of the Oxus River, one may tentatively 
assign Antimachus as a ruler in that area. 14 Later the Sogdians imitated coins of 
Euthydemus while other 'barbarian* coins copied issues of Eucratides or Heliocles. 15 
The first losses of territory of the Greco-Bactrians north of the Hindukush were 
probably to the Parthians, as we have seen, but either under Eucratides or his successor 
Heliocles, Sogdiana and other northern areas were probably lost to nomadic invaders 
sometime about 140 B.C. According to some numismatists the northern provinces of 
the Bactrian kingdom became independent earlier, beginning after the death of 
Euthydemus, since the barbarous copies of his coins with debased Greek legends are 
supposed to have been minted shortly after the death of Euthydemus. 16 Since coins of 
both Euthydemus and Heliocles have been found in Central Asia as well as copies of 
both (see note 15) it would seem that the copies were made later, even after the fall of 
the Greco-Bactrian kingdom, rather than during the reigns of the two kings. There is 
no evidence for an independent state in Sogdiana other than a possible kingdom under 
a sub-king or vassal of the main Greek ruler of Bactria, for the coins which are copies 
of the two rulers mentioned above are generally assigned to the century after the fall 
of the Greco-Bactrians. 17 In any case, we have no basis on which to determine the 
time of the falling away of the northern areas of the Bactrian kingdom, but the date of 
the famous statement of Strabo (XI, 511) listing the nomads who took Bactria from 
the Greeks, which we shall discuss below, is generally considered to apply to the time 
span 135-130 B.C., so northern areas must have been lost earlier. 

After the Bactrian Greeks moved south of the Hindukush they soon extended their 
sway over the northwestern plains of the sub-continent. It is now assumed that the 
great variety of coins were struck at a few mints, at Alexandria, near present day 
Charikar, at Pushkalavati, probably present Charsada, and at Taxila, with later mint 
sites at Gardiz and elsewhere. 18 As mentioned, Demetrius I began the conquest of 
Indian territories, but the extent of his conquests is still much disputed. The 
proliferation of bilingual coins with Greek legends on one side and a Prakrit legend in 
the Kharoshthi script on the other indicates a kind of duality of the rule of the Greeks 
to the south of the Hindukush, and Narain may be followed in designating the rulers 
after Heliocles, the last Bactrian ruler, as Indo-Greeks. Before turning to the most 
famous of the Indo-Greek rulers, it should be noted that between Demetrius I and 
Menander we must account for a number of kings, or possibly sub-kings, who for the 
most part seemed to have ruled on both sides of the Hindukush. Of the Euthydemids, 

14 Cf. B. Kastalskii, "Neizdannaya Greko-Bak- monety, chekanennye po tipu tetradrakhm Ge- 

triiskaya tetradrakhma-medal Antimakha I, bitaya liokla," EV, 11 (1956), 63-75. The latter were 

v chest Evtidema I," VDl, 3^ (1940), 347. A found in Ferghana. See also M. Mitchiner, The 

silver and a copper coin of Eucratides have been Early Coinage of Central Asia (London, 1973), 26- 

found in excavations in Khwarazm but no other 35. 

coins with Greek legends; cf. B. 1. Vainberg, ' 6 Mitchener, op. c it. [n. 1 5], 26, with references 

Monety Drevnego Khorezma (Moscow, 1976), 176. to articles of E. Drouin. 

,5 Cf. A. N. Zograf, Monety "Geraya" (Tash- l7 See Staviskii, op. cit. [n. 5], 159-60 and 

kent, 1937), 24, and Thompson, el al., supra, 239-42. 

Inventory, 265. On copies of Eucratides coins see A. l8 Cf. G. K. Jenkins, Ancient Greek Coins 

N. Mandelshtam, "K predkushanskomu chekanu (London, 1972), 261, and for Gardiz see Bivar, 

Baktrii," EV, 17 (1966), 85-90, and on copies of "Sequence" [n. 11], 129. 
Heliocles see V. M. Masson, "Drevne baktriiskie 



186 Chapter VII 

from the coins one might distinguish, as mentioned, a second Euthydemus as well as a 
second Demetrius but they are disputed. Antimachus and Pantaleon also have been 
mentioned, followed by Agathocles. Agathocles' coins are especially noteworthy in 
having Prakrit legends in the Brahmi alphabet, as did Pantaleon. But more striking 
are the 'commemorative' coins of Agathocles, a better term for these coins than Tarn's 
'pedigree' coins. Agathocles struck coins in honor of Euthydemus, of Demetrius and 
Pantaleon, as well as Alexander, Antiochus and Diodotus. 19 Then we have an 
Apollodotus who has caused controversy because of the variety of the monograms on 
his coins implying a very long reign. In such cases numismatists postulate a second 
ruler with the same name, and such is the case with Apollodotus. 20 Overstrikes on 
coins usually are good evidence for one ruler immediately succeeding the one whose 
coins are overstruck with the name of the former, but it is also possible that the 
interval in time is much greater than a few years. The various surmises on family 
relationships are difficult to resolve if based only on coins, as they are; for example, 
the queen Agathocleia, portrayed on coins with Strato, has been considered the 
daughter of Demetrius I by Tarn (p. 225), or a daughter of Demetrius II by Naraki 
(pp. 75, 181), or as a daughter of Agathocles by Simonetta (infra, n. 9, p. 160). Many 
agree that she was the wife of Menander and mother of Strato, although Simonetta 
argues that she was the wife of Apollodotus I and mother of Strato. Bivar, however, 
places Strato much later and suggests that Agathocleia was his consort rather than his 
mother, and his arguments are a bit more convincing than previous surmises. 21 

Menander, no matter who was his queen or his son, was that Greek king who made 
such an impression on the Indians by his conquests in the sub-continent that he was 
remembered in Indian sources, especially the Buddhist work in Pali the Milindapafiha, 
the questions of King Milinda, presumably Menander. It has been shown that the 
earliest name for the Greeks was a Prakrit form Yona borrowed from Old Persian 
Yauna and then Sankriticized to Yavana, with a later Prakrit form Yonaka, all usages 
prior to the Greco-Bactrian invasions of India. 22 So the names of the invaders of the 
sub-continent in various Sanskrit works are neither earlier nor later than the Prakrit 
forms. The examples of the imperfect tense referring to a recent action in the Sanskrit 
commentary of Patafijali on the grammar of Panini indicate that the Yavanas at some 
time were besieging Saketa and Madhyamika both deep in India. In Buddhist 
tradition Menander became a Buddhist hero and the prototype of a wise king. 
Whether an original Greek book called the 'Questions of Menander' of a Greek 
comedian was conflated with legends about the Greco-Bactrian king in the Buddhist 
work, as Tarn proposes, presents problems, but this is not germane to our task, which 
is restricted to the Iranian cultural area. References to Greeks and to Menander in 

19 H. P. Francfort, "Deux nouveaux tetra- 2I Bivar, "Indo-Bactnan Problems" [n. 11], 94, 
drachmes commemoratifs d'Agathocle," RN, 17 102. Altheim follows Tarn. On the sequence of 
(1975), 19-22, and P. L. Gupta, "Three Com- rulers see also K. W. Dobbins, "The Sequence of 
memorative Tetradrachms of Agathocles," JNSI, Bactrian Coins," Numismatic Digest, 2 (London, 
38 (1977), 92. 1978), 1-13. 

20 Simonetta, op. cit. [n. 9], 159. Likewise the 22 C. Tottosy, "The Name of the Greeks in 
murder of Eucratides by his son, according to Ancient India," AAH, 3 (1956), 301-18. 
Justin (XLI, 6), has been attributed to a certain 

Plato, or to Eucratides II, or to an unknown son of 
Demetrius, all guesses. 



Greco-Bactrians, Sakas and Parthiatts f #7 

Indian sources have been discussed by Narain and others, and his view on the limited 
conquests of the Greek kings in the Ganges basin, rather restricted to raids, and no rule 
of the Indus delta is convincing. The realms of the Greeks were certainly in the 
northwest of the sub-continent, and we assume the existence of a number of 
kingdoms, the changing boundaries of which cannot be determined. The 
contemporary nourishing of several kingdoms seems a more likely hypothesis than 
that of a large, unified kingdom with sub-kings as has been proposed by some 
numismatists. 

Menander is also the only Greek king mentioned in an inscription in India, and this 
is a Kharoshthi Prakrit inscription on a steatite casket found in Bijaur, north of 
Peshawar, a Buddhist dedication in the reign of the maharaja Menander. 23 
Unfortunately this tells us nothing other than the name of a later donor of the casket, 
and the reason for the name Menander remains obscure. The dating of Menander has 
raised much controversy, but the consensus now is that he ruled c. 155-140 B.C. in 
very general dates. 24 Bivar (infra, n. 24, p. 126) claims that 1 55 B.C. is not only the date 
of the accession of Menander but also the beginning of a new era of dating in ancient 
India, which may be a good guess. Also, as he suggests (p. 133), "there is a hint in one 
of the scarce literary sources (Plutarch, Moralia, 821 D) that the kingdom was divided 
after his death, and the same deduction can be made from his coins." The geographical 
location of the realms of his successors Zoilus, Lysias, Antialcidas and Theophilus, as 
well as others, cannot be determined any more than their dates, and one can say little 
other than the probable continuation of the Euthydemid rulers and either an 
extinction of the Eucratidid house or at least a drastic shrinking of any realms they 
ruled in the Hindukush mountains. As Bivar suggests (infra, n. 24, p. 173), only 
through a study of the legends, including their places on the coins, can the chronology 
of the kings be attempted, together with a review of the mongrams which give 
evidence of the place of striking, even if they are considered the signs of mint masters 
rather than the mints themselves. Further hoard finds such as the remarkable one 
from the Qunduz area in Afghanistan can revise the surmises built over the years on 
those coins previously available for study. 25 The new finds in that treasure, probably 
buried a short time after the fall of the northern Greek kingdom(s), indicate that a 
number of late rulers who hitherto had only struck bilingual coins also had struck 
coins only in Greek and in the Attic weight, such as Lysias, Theophilus, Archebius, 
Philoxenus, Amyntas and the very last king Hermaeus, while new examples of 
Antialcidas were also found (infra, n. 25, p. 61). Further, the important hoard of 
Qunduz seems to prove that the Greek coins of the Attic standard were intended to 
circulate north of the Hindukush, whereas the bilingual coins on the Indian standard 
were current to the south, although some late 'Greek only' legends may be simply 
prestige issues. A further conclusion from this hoard is that a Greek kingdom seems to 

23 N. G. Majumdar, "The Bajaur Casket of the Simonetta, op. cit. [n. 9] 162, believes Menander 
Reign ofMenmder," Epigraphica Indica, 24(1937), ruled circa 130 B.C. and not early as the others 
1-8, and comments by S. Konow in El, 27 (1940), affirm. 

52-58, as well as in the New Indian Antiquary, Jan. 25 R. Curiel et G. Fussman, Le trhor monitaire de 

1940, also his article in JRAS (1939), 265. Qunduz, MDAFA, 20 (1965), with bibliography 

24 Bivar, "Sequence" [n. 11], 126, 132; cf. of other studies. The list of rulers who struck coins 
Altheim, op. cit. [n. 1 ], 591 , and Narain, supra, Indo- with only Greek legends, only bilingual, or both 
Greeks, 181, who extends his dates to 130 B.C. types (pp. 81-82) is especially instructive. 



188 Chapter VII 

have continued to exist in the northern foothills of the Hindukush even after the 
demise of the Greek domains in India or on the Bactrian plains, after the nomadic 
invasions. The hoard is a major source for a discussion of the end of Greek rule in 
Central Asia and Afghanistan. 

We already have noted the copies of coins of Euthydemus, Eucratides and Heliocles 
in Central Asia, probably dating from after the nomadic invasions of that area or 
from the fall of the Greco-Bactrians. It is at once clear from coins and artistic remains 
that Greak influence in general was very small in Khwarazm, more in the Zarafshan 
valley, but less in the adjoining Ferghana valley, and most in the lands south of the 
Hissar mountain range, in northern Bactria south to the Oxus, and in southern 
Bactria from the Oxus to the Hindukush. This conclusion is also based on the use of 
the Greek alphabet in Bactria, whereas to the north Aramaic continued in use until 
native Khwarazmian and Sogdian began to be written in the post-Greek periods. It 
can also be seen in the later art in the different areas that Greek influences were 
strongest in Bactria, and less visible to the north, which is understandable, since they 
were far removed from the central Greek influences in Bactria. 

In the mountains of Afghanistan the Greeks seemingly continued to live after the 
nomads occupied the plains, but the city of Ay Khanum was not destroyed; by 
nomads, as far as archaeology can tell us; rather the inhabitants seem to have left 
the site sometime towards the end of the second century B.C. 26 The easiest 
interpretation of the end of Greek rule north of the Hindukush is that the rulers left 
for the mountains in the face of nomadic invasions, and the local settled people 
remained in the various towns paying allegiance to new nomadic masters, but 
allowing the Greek theater at Ay Khanum, for example, to be used as a depository for 
the bones of the dead according to local custom. Altheim has proposed that the Greco- 
Bactrian kingdom north of the Hindukush did not fall to nomads, but rather to 
Mithradates I of Parthia about 139 B.C., but then the Greeks regained their 
independence after the death of Mithradates in 138/7 B.C. to finally succumb to the 
nomads. 27 Tarn (pp. 222-23) previously had suggested that Mithradates I had 
occupied most of Bactria after defeating Eucratides, and, following a late Christian 
Latin author Paulus Orosius, suggests that Mithradates' conquests extended into India, 
but which part of 'India* is not mentioned in Orosius. It is conceivable that 
Mithradates, after taking the area of Herat from the Greco-Bactrians did campaign to 
the south in Seistan and to the east in Arachosia, for there is no coin evidence of 
continued Greco-Bactrian rule in Seistan, whereas in Arachosia the absence of early 
Parthian coins, either from excavations or in the bazaar of Qandahar is noteworthy. 28 
Since copper coins of later Indo-Bactrian rulers are relatively plentiful in this area, one 

- 6 Cf. P. Bernard in his report of the 1 976-77 28 On the question of Mithradates' conquests to 

excavations in CRAl, (Paris, avril-juin 1978) 450, the east see E. Herzfeld, 'Sakastan,' AMI, 4 (1932), 

and in the unpublished proceedings of the Kushan 40-41, and Daffina, supra, Immigrazione, 41-43, 

congress in Kabul in Nov. 1978, and in J C. who concludes that Mithradates I did not go 

Gardin et P. Gentelle, "Irrigation et peuplement farther than Arachosia. The use by Altheim of 

dans la plaine d'Ai Khanoum," BEFEO, 63 (1 976), Diodorus (XXXIII, 20) who mentions an Arsacid 

97. king (Mithradates?) who became lord of Porus' 

Altheim, Stiehl, op. cit. [ch. 6, n. 49], 597-98, realm without fighting, is rightly rejected by 

not accepted by Narain and others. Daffina and by Tarn, op. cit. [ch. 1, n. 23], 524. 



Greco-Bactrians, Sakas and Parthians 189 

may presume a raid by Mithradates rather than any lasting conquest. 29 In any case, 
the Parthians as well as the Greeks were to feel the force of the nomadic invasions 
which changed the face of the east even though the legacy of their predecessors was to 
continue in many ways. 

We know nothing of the administration of the Greco-Bactrian domains and may 
assume that Seleucid models were followed with provinces and subdivisions of 
provinces. An innovation seems to have been the 'meridarch' mentioned in several 
Indian Prakrit inscriptions, although the title was used in Phoenicia and Palestine 
under the Seleucids seemingly as a financial post. 30 There may have been a system of 
sub-kings as some numismatists have proposed, but the overlaps of kings' names on 
coins could be attributed to territorial divisions or separate domains, or even to joint 
rules. We simply do not know, but since sub-kings are not found elsewhere in the 
Hellenistic world there is no compelling reason to assume them here. Commemora- 
tive coins and overstrikes, it should be noted, are not usually taken to be evidence for 
sub-kings. In India, local divisions probably were followed, but again we have no 
sources for local administration. 

It is not possible here to discuss the many fascinating details of the legacy of the 
Greeks in Central Asia, especially in the realm of material culture, for their 
contributions to art and archaeology are continually revealed in excavations. Scholars 
agree that under the Greco-Bactrians urbanization greatly developed in Central Asia 
with the copying of well-organized square or rectangular grid plans for towns and in 
spite of hyperbole, when both Trogus (in Justin) and Apollodorus (in Strabo) call 
Bactria the 'land of a thousand cities,' we may presume there was a flowering of urban 
settlements. High walls and towers surrounded the towns and trade and commerce 
developed perhaps more, relatively, than agriculture, although increased irrigation 
also extended the cultivated land, not only in Bactria proper but probably elsewhere 
too. 31 In a series of articles and books Pugachenkova has studied the Hellenic 
influences on Bactrian architecture, especially the popularity of the Corinthian stone 
column, or modifications of it in Central Asia. 32 Other than Ay Khanum, only a 
small site called Saksan-Okhur in Tajikistan on the banks of the Panj River is dated to 
the late Greco-Bactrian period, primarily because of the quadratic plan and the central 
place of a court, as at Ay Khanum, a distinctive feature of Bactria but unlike Greek 
cities in the west. 33 So already local influences assert themselves in the time of Greek 
rule, to develop further in later periods. 

29 D. W MacDowall and M. Ibrahim, "Pre- bibliography in notes see G. A. Pugachenkova i L. 
Islamic Coins in the Kandahar Museum," Afghan I. Rempel, Isloriya Iskusstvo Uzbekistana (Moscow, 
Studies, 1 (1978), 67-68. 1965), 35-101, her Khalchayan (Tashkent, 1966), 

30 For references to the inscriptions see Narain, and "Zodchestvo antichnoi Baktrii-traditsii l 
supra, Indo-Greeks, 95, and for the west svyazi," in Pichikyan, supra, note 2, 217-24, and 
Morkholm, op. cit. [ch. 6, n. 93], 108-09. other articles in this work. Ionic columns have 

31 Cf. Gardin/Gentelle, op. cit. [n. 26], 97. They been found in Taxila, but apparently they were 
stress, however, that the population of Ay very rare compared to Corinthian capitols. 
Khanum was small even though the city walls 33 On Saksan Okhur see B. A. Litvinskii i Kh. 
were imposing, and conclude that the city was a Mukhitdinov, "Antichnoe gorodishche Saksanok- 
center for a large and well-populated surrounding hur," SA, 2 (1969), 160-78, and Litvinskii, 
area based on agriculture. "Drevnii sredneaziatskii gorod," Drevnii Vostok, 

32 For a survey of both art and architecture with Goroda i Torgovlya (Erevan, 1973), 121-24. 



190 Chapter VII 

From the coins as well as statues and other art objects some idea of the various cults, 
Greek as well as ancient Near Eastern or local Central Asian, which existed in Bactria 
may be inferred. For example, the discovery of a silver medallion with Cybele on a 
chariot at Ay Khanum suggests some attention to ancient Near Eastern cults, while a 
bronze statuette of Herakles attests the popularity of this Greek hero, even far to the 
east of the Hellenistic world. 34 Local Bactrian cults of a mother goddess and other 
deities are attested by clay figurines, and one may suppose in this period already 
considerable syncretism, or rather an identification of Greek with Iranian deities, such 
as Apollo with Mithra, although more evidence is available for the Kushan era than 
from the earlier periods. 35 Some scholars place great weight on the evidence of 
burials from archaeological sites, not only for interpretations of religious beliefs and 
practices, but also as an aid in the understanding of the society and culture of an area. 
Whether ancient Greek practices clashed with local observances is difficult to 
determine. At Ay Khanum an excavated necropolis revealed two types of burial; in a 
mausoleum both mud brick sarcophagi as well as funerary jars with bones in them 
were found, and later, presumably after the departure of the Greeks from the city, 
many bones were found all over the stage part of the Greek theater. 36 The exposure of 
bodies to birds and animals is a well-known Zoroastrian custom, if not an even rnore 
ancient eastern Iranian practice, so the discovery of such a way of disposal of the dead 
at Ay Khanum should not be surprising. What we cannot determine is how the 
different modes of disposal of corpses co-existed, and if there was any interchange or 
adaptation in practices by one segment of the population with another. 

We may conclude that the almost two centuries of Greek rule in Bactria and 
Central Asia saw a concentration of Hellenistic settlements and the founding of cities 
in Bactria proper, with a conservativism in the preservation of Greek customs, the 
Greek language, and a separation between the Greek rulers (by Greek also 
Macedonian, Thracian and Anatolian is implied) and the local population. There is no 
evidence of serious conflicts between the Greeks and the local population; on the 
contrary, a gradual fusion of the two seems to have progressed. With the 
independence of the Greeks in Bactria we may presume that colonists from the west 
ceased completely. Even if Eucratides were found to be a Seleucid general sent by 
Antiochus IV to reassert Seleucid rule in Bactria, which is unlikely, he surely did not 
bring a contingent of Greek colonists with him. Most likely after the death of 
Antiochus III any Bactrian connections with the Seleucids were only on the basis of 
agreement or treaty between two sovereign states. It is possible that an exchange of 
mercenaries existed at times between the two states, but any consequences of such 
hiring for both powers were negligible. An impressionistic but colorful assessment of 
the legacy of Alexander in the east has been provided by Wheeler, while popular 
books in Russian on the general subject abound. 37 The main difference in our 
knowledge of the heritage of the Greeks today as compared with half a century ago is 
the end of what was called the 'Bactrian mirage,' with a realization from finds at Ay 

34 For illustrations and references see Allchin/ 37 M. Wheeler, Flames over Persepolis (London, 
Hammond, op. cit. [ch. 6, n. 43], 227. 1968), 63-122, and in Russian, V. M. Masson, 

35 G. Pugachenkova, "O kultakh Baktrii v svete Slrana tysyachi gorodov (Moscow, 1 966), 33-73, B 
arkheologie," VDI, 3 (1974), 125-35. Ya. Staviskii, Mezhdu Pamirom i Kaspiem (Mos- 

36 Bernard in CRA1 (avril-juin 1978), 440-41. cow, 1966), 78-95, and many others. 



Greco-Bactrians, Sakas and Parthians 191 

Khanum and from Soviet excavations in Central Asia that the later Buddhist art 
called Gandharan did have roots in Bactria, even though it was a revival based on 
impulses from the Roman Empire. New coin hoards and finds of Greek coins in 
excavations will enable numismatists to better establish geographical and chronologi- 
cal frameworks for the many rulers of this part of the world in the two centuries 
before our era. 



NOMADIC INVASIONS 

This period of history is generally called that of the Saka and Indo-Parthian dynasties 
previous to the rise of the Kushan Empire. For the first time, Chinese sources may be 
utilized with caution for a reconstruction of history in this part of the world. A 
powerful Chinese emperor Wu ti of the Han dynasty sent an emissary to the west 
called Chang Ch'ien who returned to China in 126 B.C. His report was included in the 
history Shih-chi, chapter 123, which account essentially is repeated with many 
embellishments in the Han shu, or annals of the former Han dynasty. 38 The Chinese 
accounts have been translated so many times it almost seems superfluous to repeat 
them, especially as there are no great problems in the translations as there are in the 
interpretations. The Shi-chi simply says that the Yueh-chih were defeated by the 
Hsiung-nu and went far away. They passed Ta-yUan and in the west they attacked and 
subjugated Ta-hsia. They made their capital north of the Kuei-shui (Oxus River) and 
made it their royal court. 39 Much has been written about the Chinese ethnic and 
geograhical designations, especially about the name YUeh-chih which has been 
identified as a Chinese transcription of larioi, of Tokhar(ian), of Scythian or as the 
great 'Ghara' people or 'mountaineers.' 40 Whatever the etymology, they should be 
identified as one of the nomadic tribes which invaded Bactria, according to 
Apollodorus (Strabo) and Trogus (Justin). The former names the Asii, Pasiani, 
Tochari and Sakarauli, while the latter says the Saraucae(sic) et Asiani seized Bactria 
and the Sogdians. 41 The name Sacaraucae has been explained as 'Saka commanders' by 
Bailey, and they may be identified as a tribe of the 'royal Scythians' or Sai-Wang 'Saka 
king(s)' of Chinese sources. 42 Since many tribes in the steppes in later times had clans 

38 See E. G. Pulleyblank, "The Wu-sun and 4I Pompeius Trogus, Fragmenta, ed. by O. Seel 
Sakas and the YUeh-chih migration." BSOAS, 33 (Leipzig, 1956), 178, book 41. The enigmatic 
(1 970) 1 54-60. caption at the end of the summary of his 'addition 

39 Ibid., 154, and Narain, supra, Indo-Greeks, to Scythian affairs,' reges Tocharorum Asiani interi- 
129. tusque Saraucarum, book 42 on p. 180, has been 

40 H W Bailey proposes an Iranian form of the emended in various ways. The last name is 
Chinese Yueh-chih as *yah-iik and associates the obviously a slip for Sa(ca)raucae. For another view, 
name *Tokhara with the Chinese designation as and an etymology of Sacaraucae, see R. Schmitt in 
the Ta or 'great Ghara' people. Cf. his "Sri Visa Sura WZKM, 67 (1 975), 87. 

and the Ta-uang," AM, 11 (1964-65), 6, with 42 H. W. Bailey, "North Iranian Problems," 

further references. Bailey returned to his Ghara BSOAS, 29 (1979), 27-28. For a criticism of 

theory in his lectures delivered at Columbia Haloun's identification of the word YUeh-chih as 

University, N.Y. in the fall of 1979 Pulleyblank, 'Scythian,' see E. G. Pulleyblank, "Chinese and 

op. cil. [n. 38], and in his "The Consonantal System Indo-Europeans," JRAS (1966), 17. The Asii 

of Old Chinese," AM, 9 (1962), 90, however, generally have been identified with the As-Alans, 

proposes the identification Tokhanan = Ta-yUan, the descendants of whom are the Ossetes of the 

which is usually identified as Ferghana north Caucasus who speak a neo-'Saka' language. 



192 Chapter VII 

or sub-tribes which had prerogatives of rule, it is not unlikely that much earlier the 
nomads also had a royal clan to which others owed allegiance. The name Pasiani has 
been called a misreading of Parsi 'Persians' or simply a repeat of Asii with an Iranian -n 
ending and the Greek feminine article replaced by a p-, but there is little one can do 
with the text. 43 Since it is not relevant to our history whether the Ytieh-chih 
originally spoke a dialect of the centum Indo-European language called Tokharian, the 
problems of identification of words and etymologies are not of concern to us. 

It is significant that according to the Classical sources several groups of nomads 
invaded Bactria, while all of the Chinese sources speak only of the Yueh-chih, 
although the early Han shu adds the information that in the previous course of their 
movement in the north the Yueh-chih had defeated the Sai-wang who then went 
south. 44 The easiest way to reconcile the two is to assume that the Chinese mentioned 
two large but different peoples, the Yueh-chih and the Saka, whereas the Classical 
sources spoke of tribal names, the Asii or Asiani, being a Saka tribe, and the 
Tokharians either to be identified with the Yueh-chih, or at least part of them. Ta-hsia 
has been identified as the land of the Tokharians, as Greeks, as Daha to the east of the 
Caspian and otherwise, but whatever the phonetic identification or meaning, Ta-hsia 
should be either the Bactrian kingdom or a part of it. Over the years Sinologists have 
wrestled with Chinese transcriptions of foreign words and many enigmas still 
remain, and given the Chinese propensity to designate peoples or tribes as Ta 'large' or 
Hsiao 'small,' as well as the genius of the language for puns, many mysteries, such as 
identifications of western tribes, will remain to plague the reconstruction of the past. 
From both the Classical and Chinese sources we may say that sometime in the second 
century B.C., and estimated dates have ranged from 175 to 150 B.C., the Yueh-chih 
moved from the western borders of China under pressure from the Hsiung-nu across 
Sinkiang, into the territory of the Sakas in the Hi-Ferghana area. Some of the §akas 
then went south about the middle of the second century B.C. and the question arises 
whether these Saka had anything to do with the fall of Bactria. Narain (p. 134) 
suggests that the Sacaraucae and Asianii, two Saka tribes, as a result of being displaced 
in their northern homeland by the Yueh-chih, were those nomads who caused 
trouble for the Parthians after the death of Mithradates I. These western Saka he 
distinguishes from eastern Saka who moved south through the Kashgar-Tashkurgan- 
Gilgit-Swat route to the plains of the sub-continent of India. This would account for 
the existence of the ancient Khotanese-Saka speakers, documents of whom have been 
found in western Sinkiang, and the modern Wakhi language of Wakhan in 
Afghanistan, another modern branch of descendants of Saka speakers parallel to the 
Ossetes in the west. It also accounts for the large number of coins of Saka rulers found 
in those northwestern regions of the sub-continent. Thus, according to this theory, 
the Sakas moved south on both sides of Bactria but did not overthrow the Bactrian 

43 For a survey of suggestions on emendation see of wang 'king' as representing the last part of the 
Narain, supra, Indo-Greeks, 1 32. Daffina, supra, word Sacaraucae, and reviews other theories. The 
Immigrazione, 54—56, equates the Pasiani with the existence of a tribe or sub-tribe of 'Royal Scythians' 
Apasiakai, which does not help us with further is not in doubt, only the Chinese usage, and it is 
deductions. difficult to decide if the passage means that the 

44 Narain, supra, Indo-Greeks, 130-31. Daffina, 'king of the Saka' moved south or the Sacaraucae 
supra, Immigrazione, 46, disputes the identification moved south 



Greco-Bactrians, Sakas and Parthians 193 

kingdom, which was left to a later migration of the Yileh-chih. From the Chinese 
sources one would assume that the decline and fall of the Bactrian kingdom was not a 
swift but rather a gradual process perhaps in several stages. 

Beyond the mention of the migrations in the Chinese and Classical sources we have 
no other literary sources and archaeology cannot fill the gap. Excavations o£kurgans 
or burial mounds of nomads from this period have given one indication which is the 
close similarity in the styles of objects left by those nomads who invaded Bactria and 
the art of the Sarmatians of South Russia, of whom the Alans-As were one tribe. 45 
Thus archaeology does coincide with the sparse linguistic data to confirm the view of 
the nomadic invasions of Bactria as part of the great movement of tribes over Eurasia 
in this period. The theory that the Ytieh-chih were the speakers of the 'centum' Indo- 
European tongue called Tokharian by contemporary linguists is possible but 
relatively unimportant historically, since they adopted the local Iranian language of 
Bactria as their literary language and were in any case Iranicized. One theory had the 
Sacaraucae and Asiani first conquer Bactria to be followed after a decade by the Yueh- 
chih or Tokharians. The Sacaraucae moved south, but the Asiani remained in Bactria 
and became feudatories of the Tokharians, and it was the Asiani who later supplied 
the chiefs of the Kushans. 46 This view is also based on ch. 123 of the Shih-chi, which 
suggests that even though the Yueh-chih conquered Ta-hsia (Bactria) the former 
maintained their court north of the Oxusand left the latter more or less independent. 
All that may be said is that from archaeological excavations and from the absence of 
literary sources there is no evidence of violent destruction of Bactrian cities on the 
part of the nomads, and one may presume some sort of co-existence between the 
nomadic and settled population over a period of time, but any details are lacking. 47 

The Saka obviously went south to Drangiane or Zranka and gave their name to 
this province, Sakastan or Seistan today, but when this happened is disputed. If they 
followed a usual nomadic pattern rather than one of an invading army, the Sakas, 
whatever tribes participated in the move to the south, probably infiltrated the 
countryside from Merv to Herat and Seistan in the time of Mithradates I. It was under 
his successor Phraates II that the Saka became belligerent and defeated him. So in the 
middle of the second century B.C. the Saka were moving south into Seistan but 
probably not much beyond. Since two Parthian kings, Phraates II (died c. 128 B.C.) 
and Artabanus II (died c. 123 B.C.) lost their lives in warfare against the nomads, we 
may assume that in their time Saka power was at its height and they had occupied 

45 See esp. A. M. Mandelshtam, Pamyatniki 46 L. Bachhofer, "On Greeks and Sakas in India," 

Kochevnikov Kushanskogo Vremeni v Severnoi JAOS, 61 (1941), 247. 

Baktrii (Leningrad, 1975), 148; also the series * 7 Daffina, supra, Immigrazione, 65-67, pro- 

Uspeki Sredneasiatskoi Arkheologii, vypusk 1-3 posed that the Sacaraucae and the Pasiani crossed 

(Leningrad, 1972-75). The continuity of culture the Oxus more or less together, the former moving 

from this period to the later Kushan era is also south to Seistan while the latter occupied Bactria. 

striking. O. Maenchen-Helfen in "The Yueh-chih As mentioned, he identifies the Pasiani with the 

Problem Re-examined," JAOS, 65 (1945), 79-80, Apasiakai, mentioned by Strabo (XI, 513) instead 

discusses the western movement of the Asii-Asiani- of considering the form with P- a mistake. Both 

Aorsi-Arsi, and he asserts these were lords of the guesses are possible, but, as usual for the east, we 

Tokharians, while the Digor dialect speakers of must say non-liquet. 
Ossetic are descendants of the Tokhari, an interest- 
ing but unsubstantiated surmise. 



194 Chapter VII 

some of Bactria as well as eastern Iran. The relationship between these Saka and the 
Yileh-chih to the north is unknown, but no evidence exists for any relations in the 
second century B.C. What happened to the Saka in eastern Iran after the defeat of the 
Parthians? 48 Presumably the beginning of a Saka kingdom in Seistan may be dated 
from this period. 

One scholar has maintained that there was only one Saka movement in this period 
of the Yileh-chih invasion of the Bactrian kingdom (c. 1 30 B.C.) and this was an 
eastern invasion of Kashmir and the northwest part of the Indian sub-continent, 
following the Chinese annals of the former Han dynasty, from Sinkiang. 49 This 
movement of one group of Sakas, however, need not obviate the need to explain 
Seistan, which was hardly settled by Sakas coming from India. The fact that the 
Parthian Stations of Isidore of Charax (17 and 18) mentions both 'Zarangiane' and to 
the east of it Sakastane, implies the settlement of Sakas to the east of the Hamun lake 
by the first century B.C., since Isidore flourished in the following century. 50 If a Saka 
state were created in eastern Iran, it did not maintain its independence long, for 
Mithradates II (c. 123-87 B.C.), as Justin (XLII, 2) informs us, added many areas to the 
Parthian kingdom and several times fought successfully with the Scythians, and 
avenged the previous Parthian defeats. There is no evidence that Mithradates II either 
restored the limits of conquest of his predecessor Mithradates I or that he exceeded 
such boundaries in the east. All that may be said is the implication of Isidore's text that 
Parthian rule included Seistan and Arachosia, although clearly at times one or both 
areas were either fully independent or in some sort of dependent relationship with the 
Parthian kingdom. Before turning to the Parthian expansion in the east, however, the 
Sakas in India and the last of the Greco-Bactrians deserve attention. 

The Saka migration to India from Sinkiang, part of the general movement of 
peoples in Central Asia in the second century B.C., has been mentioned, but a 
chronology and geographical details are lacking. The annals of the former Han 
dynasty (chs. 61 and 96) say that the Sai (Saka)wang went south and subdued Chi-pin; 
the Saka tribes were scattered and constituted several kingdoms in various 
countries. 51 Much has been written about the identity of Chi-pin, although all agree 
it is a land to the south of the Hindukush-Himalaya mountain range, some 
identifying it with Gandhara, others with Kafiristan and still others with Kashmir. 
Pulleyblank restores the ancient pronunciation as kiei-pyin < *ka(t)s-pen = *Kaspir 



48 J- Junge, Saka Studien (Leipzig, 1939), 102- him, this accounts for the existence of the Khotan- 

03, following G Haloun, "Zur Oe-tsi-Frage," Saka language and close connections with India. 

ZDMG, 91 (1937), 256, emends Asiam in the text 50 On Isidore's dates see Dafrina, supra, lmtni- 

to *cusani and interprets interims as the destruction grazione, 72, 75, 82. 

of Saka power in India by the Kushans. Tarn, op. 5I Translation in Narain, supra, Indo-Greeks, 

cit. [ch. 1, n. 23], 306, explains this word by the or in Bachhofer.op. of, [n. 46], 242, in J. de Groot, 

defeat of the Saka near Merv by a Parthian of the Chinesische Urkunden zur Geschichle Asiens II, Die 

Suren family who 'drove them across the Oxus Westlande Chinas (Berlin, 1926) 86-91, and 

where they perished,' all of which is unconvincing. Altheim, op. cit. [ch. 6, n. 19], 608. Variations in 

Cf. Damna, supra, Immigrazione, 77-80, who translations are not vital to the general information 

accepts the view of Junge. that some Saka moved south to India, and there is 

Junge, op. cit. [n. 48], 105-06. According to no reason to doubt this information. 



Greco-Bactrians, Sakas and Parthians 195 

for Kashmir, which it probably was at a later period, but perhaps also in the second 
century B.C. as well. 52 In any case, some Sakas, but maybe only a few, crossed the 
mountains to the sub-continent of India. This account has been disputed by Tarn (p. 
321), Daffina (supra. L'Immigrazione, 47) and Altheim (p. 609), who claim it is a later 
addition to the original text, when the Sakas from Seistan had established kingdoms in 
India, and the southward migration was a mere invention, since for them all of the 
Sakas in India came from the west. The first Saka ruler in northwest India was Maues, 
called Moa in the Kharoshthi alphabet on his coins, or possibly the same as the name 
Moga in a Kharoshthi inscription on a copper plate from Taxila. 53 There is no other 
information except the coins, and numismatists agree that the centre of Maues' rule 
was Taxila and territories to the north. One feature of the Taxila copper plate 
inscription raises questions, for it has the date 78, in an unknown era, while other 
inscriptions also with dates have produced a great literature about the existence of 
various eras in the sub-continent which were used for dating inscriptions. In India, as 
in the Seistan of Iran, we learn about the Sakas from later evidence of their presence, 
for example, in Sanskrit texts such as the epics the Ramayana (IV, 43, 12) and the 
Mahabharata (II, 32, 17), but the history of the Sakas in India with their probable 
descendants in the Western Kshatrapas and others is beyond the scope of the present 
book. 54 

Numismatists now agree that Maues must have ruled in northwest India around 80 
B.C. although the relative chronology of this ruler who had a long and varied series of 
coins is better understood than an absolute dating. The rule of early Saka chiefs in 
northwest India is closely bound with that of late Greek kings, according to 
numismatists, because of the copying of coin types and especially similar titles and the 
same monograms on the coins. It is impossible to discuss in the space here the various 
theories of dates and order of rulers in this part of the world, based on the same title on 
coins of different rulers, on overstrikes of one ruler on the coins of another and on the 
appearance or disappearance of monograms from various stages of the coinage of a 
ruler, since new finds of coins continually bring revisions to theories on the order of 
rule or the territories ruled by one king or another. One way of showing the 
changing fortunes of rule is to present a table of rulers and mints, usually identified by 
monograms on coins as well as their find spots, much of which is accepted by most 
numismatists such as Jenkins, Narain, MacDowall, Mitchiner, Mukherjee, Simonetta 



52 Pulleyblank, "Consonantal System" [n. 40], Meuakos in inscriptions from South Russia; cf. L. 
218. This identification is disputed by L. Petech, Zgusta, Die Personennamen griechischer Smdte der 
Northern India according to the Shui-ching chu ndrdlichen Schwarzmeerkiiste (Prague, 1955), 118. 
(Rome, 1 950), but others support Pulleyblank. For The etymology is uncertain. 

various views see Narain, supra, Indo-Greeks, 136. 54 Likewise fascinating topics such as the home 

53 For the coins see Jenkins/Narain, supra, Coin of the Sakas, the legendary continent of Saka-dvtpa 
Types, 1 , and for the Taxila plate inscription S. in Indian sources, the maga priests of the sun cult, 
Konow, Kharoshfhi Inscriptions, Corpus Inscrip- especially in the famous temple at Multan, the 
tionum Indicarum (Calcutta, 1929), 28-29. Com- maga-brJhmanas, and other questions belong to 
pare the name Maues with a Saka chief MauakSs at Indian history. 

the battle of Gaugamela in Arrian (III, 8, 3) and 



196 



Chapter VII 



z 













u 




z 




o 




Q 




UJ 




CO 




< 




S3 




oo 


^ 


UJ 


4J 


H 
oo 


o 


Q 


-o 


Z 


« 


< 


nj 






oo 


4J 


UJ 


C 







8 

1-1 


E 
(75 


O 


E 
-S 

J3 


z 

o 


a 


o 


i 


C 


u 


E 


w 





> 


is- 






H 




< 




_l 




UJ 




os 




U-, 




O 




UJ 




J 




ffl 




< 




H 





03 



o « 

2e i 



I E. 



^ 




(•J 


i/j 




rt 


a 




<u 


U 




> 






L) 




nj 


a. 


1! 


~ 


rs 


4J 


-n 



— ■ •-■ <^/ — r-i (>. n ™ — vi w ^ •----• 



— ! CB 



E 
> 



a s 



•o E 



I fc§ 



o 
-a 
*, o 

3 O 



S O - v B 



Z(i<2<a:i<< 



a s-s 

E o"> 
< > 



— i> 



< a 



E 3 

< Oh 



-3 « 



O 2 



c 
o 
O 



J3 

o 

c 
o 

O 



o 

-a 



o 
O 



-o 

J3 
< 



E 
> 



T3 

< 



E 
> 



-a 

-a 
< 



E 
> 



8^38 



— V 



OO 



a. a, a. I: 
oo oo oo < 



< a. 



I* J2 <* 

§ t J 1 
J" s. 

K> CO 



J3 •» -3 -i; 



a. 2 
co < 



C a. n 

•5 c -o 
>- o « 

ooo 



a. 
o 
-o 



00 



o 



) CO 



< d. 



O 



^ n *- o> 
c S C 2 

O c n ~ 



« m 



E 
> 



Greco-Bactrians, Sakas and Parthians 197 

and Dobbins. 55 General works by Narain, Tarn and Van Lohuizen-de Leeuw are not 
so useful as the detailed articles in establishing a chart of the rulers. 56 The chart below 
is obviously most tentative, since absolute chronologies in the east elude us; 
furthermore no two numismatists would agree on all dates and geographical extent of 
rules as proposed here, while Parthian or Saka vassal relationships are very difficult to 
discern simply from coins. At least from Indian sources we can distinguish between 
Sakas and Pahlavas, or Parthians, and we may assume that these two peoples, closely 
related to each other, divided the rule in the eastern part of the Iranian world. The 
relationship between Maues and Azes is unknown as is any connection between the 
Vonones group and Azes, but it is important to remember that the Sakas came into 
India probably as a tribal invasion or expansion, whereas the Parthians more likely 
came as a conquering army. Since the Parthian kingdom itself was feudal with semi- 
independent dynasts under the titulary rule of one king, it is not possible to determine 
the inter-relations and allegiances of the names we find on coins. We omit the names 
and titles of later Saka rulers in Western India, such as the Kshatrapas; their 
proliferation indicates a fragmentation of rule in the east. Likewise, titles have been 
omitted, for the pretensions of minor rulers to grandiose titles confuses attempts to 
arrange relationships according to titles, even though the changing titles do seem to 
indicate relationships between probable contemporaries. 

THE PARTHIANS IN THE EAST 

Literary sources are very sparse, and they are not unequivocal, but we have already 
mentioned probable conquests of Mithradates II in the east. Then come the Saka 
invasions. Several coins of Gotarzes I (c. 90-80 B.C.) have the names of mints written 
in full in Greek: Areia, Traxiane (present Sarakhs?) and Margiane, indicating 
Parthian rule over those areas. 57 Next we hear in Lucian (Makrobioi, 15) that 
Sinatruces, king of Parthia, was placed on the throne by the Sacaraucae when he was 
eighty years old and he then ruled for seven years. This statement implies an 
importance and power for the Sakas in the eastern part of the Parthian realm in this 
period. Thereafter we hear no more evidence of direct Parthian rule in the east, so we 
must turn to the coins, and we find first those of Vonones and his Saka successors. 58 

55 The sources for this reconstruction are same authors, plus others by A. N. Lahiri and D. 

Jenkins/Narain, supra, Coin-types; Jenkins, "Indo- Bivar, may be found in the bibliographies of the 

Scythic Mints," JNSl, 17 (1955), 1-26; D. above. 

MacDowall, "The Dynasty of the Later Indo- 56 J. E. Van Lohuizen-de Leeuw, The "Scythian" 

Parthians," NC (1965), 137-48, and his article in Period (Leiden, 1949) and a long review of it by 

JNSI, 29 (1968), 24-48; M. B. Mitchiner, Indo- A. L. Basham in BSOAS, 15 (1953), 80-97. 

Creek and Indo-Scythian Coinage, 3 (London, 57 D. Sellwood, /l/i Introduction to the Coinage of 

1976), a detailed compilation of coins; B. N. the Parthians (London, 1971), 80, and W. Wroth, 

Mukherjee, An Agrippan Source -A Study in Indo- Catalogue of the Coins of Parthia, in The Creek 

Parthian History (Culcutta, 1969); Simonetta, op. Coins of the British Museum (London, 1903), 140, 

cit. [n. 9], 154-83, and his "La monetazione Indo- n. 1. 

Partica ed il suo significato per la cronologia dei 58 The name Vonones (probably Parthian: 

Kushana," Rivista Italiana di Numismatica, 73 umwn 'star name?'; cf. F. W. K. MUller, "Ein 

(Milan, 1971), 33-47; K. W. Dobbins, Coinage Doppelblatt aus einem manichaischen Hymnen- 

and Epigraphy of the Sakas and Pahlavas (Canberra, buch," Abh.PAW [1912], 35) was held by several 

Ph.D. thesis, 1 972) to be published in Islamabad in Parthian kings, but his successors have Saka names : 

future. Until then see his book Saka-Pahlava Spalahora and Spalirises, from spada 'army' with 

Coinage (Varanasi, 1973) 193 pp. Articles by these the characteristic East Iranian d > 1. 



1 96 



Chapter VII 



o 

B 
O 



o 
OS 



D 
< 



8 



1/5 

z 

o 

u 

z 

o 

Q 
w 

1/5 
< 

CQ ^__ 

or> '2* 
tt! v 

H •£ 



9 "8 

Z « 

< 2 

g 1 



3 
_ 

U =5 3 

^ 8. 5 



E.2 
S ft- 
? o 



« 0. 5 ~ N N »- N ^ 2/2 



n T3 
y5 < 



e 
> 



o 



O ■fi 

7 S 

X Q 

<-> s 

gj O 

> !ii 
P 






O C i> =3 



- 

2- » 



o - a 

Cl. f — 



Z£<2<£X<< 



CQ 
< 



5-1 a. 

E §"> 

< > 



5-8 

J 3 

< a. 



J! 



< W 



o 
■a 



■7 e 
O 2 

"2 J= 



Jj 



* o I 

2 s.js £•.= 

5 c/5 « 3 « I 
j° a..Q-.a.J 



ft- O" &•.£ 
1/5 1/5 i/) < 



< £ 



o 
-a 
c 
o 
O 



o 
-o 

c 
o 
O 



o 
-o 
c 
o 
O 



o 

; -o 
: c 
. o 
i O 



T3 
-ft 

< 



s 
> 



T3 
< 



s 
> 



-a 

< 



E 
> 



O 



00 n 
2 J2 

" s 

O J? 



S3 



«i J2 « 

2 B.J3 

5 1/5 rt 

.2 ft- 
> 1/5 





l/l 


.e 






8 1 


a. 
o 
-a 


8 
I-, 
o 




5 J* 
< 0. 


c 
o 
O 


ft. 



o £S 



X *> c *> ^ 
s S ~ « 

£ g s s p 
J* o 

(/) 



sax 



us ^ 



E 



Greco-Bactrians, Sakas and Parthians 191 

and Dobbins. 55 General works by Narain, Tarn and Van Lohuizen-de Leeuw are not 
so useful as the detailed articles in establishing a chart of the rulers. 56 The chart below 
is obviously most tentative, since absolute chronologies in the east elude us; 
furthermore no two numismatists would agree on all dates and geographical extent of 
rules as proposed here, while Parthian or Saka vassal relationships are very difficult to 
discern simply from coins. At least from Indian sources we can distinguish between 
Sakas and Pahlavas, or Parthians, and we may assume that these two peoples, closely 
related to each other, divided the rule in the eastern part of the Iranian world. The 
relationship between Maues and Azes is unknown as is any connection between the 
Vonones group and Azes, but it is important to remember that the Sakas came into 
India probably as a tribal invasion or expansion, whereas the Parthians more likely 
came as a conquering army. Since the Parthian kingdom itself was feudal with semi- 
independent dynasts under the titulary rule of one king, it is not possible to determine 
the inter-relations and allegiances of the names we find on coins. We omit the names 
and titles of later Saka rulers in Western India, such as the Kshatrapas; their 
proliferation indicates a fragmentation of rule in the east. Likewise, titles have been 
omitted, for the pretensions of minor rulers to grandiose titles confuses attempts to 
arrange relationships according to titles, even though the changing titles do seem to 
indicate relationships between probable contemporaries. 

THE PARTHIANS IN THE EAST 

Literary sources are very sparse, and they are not unequivocal, but we have already 
mentioned probable conquests of Mithradates II in the east. Then come the Saka 
invasions. Several coins of Gotarzes I (c. 90-80 B.C.) have the names of mints written 
in full in Greek: Areia, Traxiane (present Sarakhs?) and Margiane, indicating 
Parthian rule over those areas. 57 Next we hear in Lucian (Makrobioi, 15) that 
Sinatruces, king of Parthia, was placed on the throne by the Sacaraucae when he was 
eighty years old and he then ruled for seven years. This statement implies an 
importance and power for the Sakas in the eastern part of the Parthian realm in this 
period. Thereafter we hear no more evidence of direct Parthian rule in the east, so we 
must turn to the coins, and we find first those of Vonones and his Saka successors. 58 

55 The sources for this reconstruction are same authors, plus others by A. N. Lahiri and D. 

Jenkins/Narain, supra, Coin-types; Jenkins, "Indo- Bivar, may be found in the bibliographies of the 

Scythic Mints," JNSI, 17 (1955), 1-26; D. above. 

MacDowall, "The Dynasty of the Later Indo- 56 J. E. Van Lohuizen-de Leeuw, The " Scythian" 

Parthians," NC (1965), 137^*8, and his article in Period (Leiden, 1949) and a long review of it by 

JNSI. 29 (1968), 24-48; M. B Mitchiner, Indo- A. L. Basham in BSOAS, 15 (1953), 80-97. 

Greek and Indo-Scythian Coinage, 3 (London, 57 D. Sellwood, An Introduction to the Coinage of 

1976), a detailed compilation of coins; B. N. the Parthians (London, 1971), 80, and W. Wroth, 

Mukherjee, An Agrippan Source- A Study in Indo- Catalogue of the Coins of Parthia, in The Creek 

Parthian History (Culcutta, 1969); Simonetta, op. Coins of the British Museum (London, 1903), 140, 

cit. [n. 9], 154-83, and his "La monetazione Indo- n. 1. 

Partica ed il suo significato per la cronologia dei 58 The name Vonones (probably Parthian: 

Kushana," Rivista Italiana di Numismatica, 73 wnwn 'star name?'; cf. F. W. K. MUller, "Ein 

(Milan, 1971), 33-47; K. W. Dobbins, Coinage Doppelblatt aus einem manichaischen Hymnen- 

and Epigraphy of the Sakas and Pahlavas (Canberra, buch," Abh.PAW [1912], 35) was held by several 

Ph.D. thesis, 1 972) to be published in Islamabad in Parthian kings, but his successors have Saka names : 

future. Until then see his book Saka-Pahlava Spalahora and Spalirises, from spada 'army' with 

Coinage (Varanasi, 1973) 193 pp. Articles by these the characteristic East Iranian d > 1- 



198 Chapter VII 

How can we interpret the legends on the coins? Presumably if we find the title 'king 
of kings' on an eastern coin this would indicate complete independence from the 
Parthian great king, but the use of other titles, such as BACIAEYZ MET AC, does 
not thereby imply a vassal relationship. Coins with two names, one in Greek on the 
obverse and Kharoshthi on the reverse apparently do show a subordinate relationship 
such as coins of Indravarma and Aspavarma who are called strategos the Greek word 
for 'general' in the Kharoshthi script, but with 'great king of kings Azes' in Greek on 
the obverse. On the other hand, we have 'great king of kings Orthagnes' in Greek on 
his coins, but on the reverse the same in Prakrit and Kharoshthi script with the name 
Gondophares, which has elicited many theories on the relationship of the two. 
Chance remarks in Classical or Indian sources can be enigmatic as well. We have 
mentioned the Parthian Stations of Isidore in which all the land to and including 
Arachosia is said to be under imperial Parthian rule, but the reality of such rule in the 
time of Isidore, who probably lived in the first century of our era, is dubious. 
Likewise the accounts of a king called Phraotes, who presumably ruled in Taxila, 
mentioned in the life of Apollonius of Tyana by Philostratus (II, 42), have been used 
variously to reconstruct a history of northwest India even though the moralistic and 
legendary nature of the work is apparent. 59 Another source is the Periplus of the 
Erythraean Sea (Para. 38), which notes that the town of Minnagar, near the mouth of 
the Indus River, was subject to Parthian princes who were constantly driving each 
other out of it. From this remark we may infer merely that in the lower Indus valley 
area Iranian chiefs fought with each other, but beyond this we cannot go, for Saka 
lords might well be confused with Parthians in this period by a foreigner. Indian 
sources, such as the Jain legend on the origin of an era ascribed to Vikramaditya on his 
defeat of the Sakas, are even more unreliable or unusable than the Classical sources, 
in spite of attempts to use them as historical. 60 Chronologies with precise years of 
reign of various rulers should be rejected when presented as fact; they should be 
regarded only as guesses, some obviously quite fanciful, for only a relative chronology 
on very general lines can be accepted at this time. The question of eras, however, will 
be discussed under the Kushans. 

Vonones, it seems, was the first independent ruler to issue coins, possibly in Seistan 
but more likely in Arachosia; associated with him was either his real brother or one 
who had an honorary title of just brother of the great king' as written in Kharoshthi 
on the reverse of coins of Vonones. This latter person had a Saka name Spalahora, and 
he was succeeded by his son Spalagadama, who like his father kept the name Vonones 
in Greek on the obverses of some of his coins; but on other coins, presumably later, 
we find in Greek the name Spalyrios with the title 'just brother of the king.' 61 What 

59 Herzfeld, "Sakastin" [n. 28], 113, identified South Asian Documents on the Old Saka Era 

this Phraotes as an epithet apratihata 'triumphant,' (Varanasi, 1973), 89-93, with good reasons, 
of Gondophares, which was rejected by Van " Cf. Jenkins/Narain, supra, Coin-types, 3-4. 

Lohuizen-de Leeuw, op. cit. [n. 56], 360, although The names have been given Saka etymologies by 

Tarn, op. cit. [ch. 1, n. 23], 341, accepted it. It is H. W. Bailey, "Ambages indoiranicae," Annali, 

better not to use Phraotes as an historical ruler Istituto Orientate di Napoli Sez. Ling. I, 2 (1959), 

when no coins of his have been found. 130-32. Vonones has been identified as a member 

'"As Altheim, op. cit. [ch. 6, n. 1 9], 61 7 ; cf. also, of the Parthian Suren family by Tarn, op. cit., [ch. 

but more cautiously, Tarn, op. cit. [ch. 1, n. 23], 1, n. 23], 344, and others, and this may be so, but 

335, and rejected by B. N. Mukherjee, Central and hardly his Saka successors. 



Greco-Bactrians, Sakas and Parthians 199 

do these titles mean? Attempts to explain the name of Vonones on the coins as an 
appellative for Mithradates II are based on a false etymology for Vonones 'the 
victorious' and are to be rejected. 62 The importance of the nephew of the king as his 
successor, however, is a widespread though not obligatory practice in Central Asia 
and is known from the Parthians to the Qarakhanid Turks of the eleventh century, 
to mention only two peoples where the practice of succession is attested. 63 
Unfortunately, only brief notices exist of the Parthian 'brother of the king,' and no 
conclusions can be drawn about the nature of the institution or practice, and we have 
no sources other than the coins from the east. 64 Whether the institution of 'brother of 
the king' was similar to, or the equivalent of, a later title found in Iran, 'second' in rule 
after the king, is unknown, but obviously the problem of succession must have been 
important, especially in a nomadic society where life could be precarious and any 
institution of a vice-king or even sub-king would insure continuity. 65 Thus, because 
of this alone it is extremely difficult to reconstruct a history of the rulers in the east 
solely from coins. 

It seems that after the rulers Vonones and his true or honorary brother Spalahora, 
followed by Spalagadama and then Spalyrios, who may have been the brother of 
Spalagadama, we have a certain Spalirises, who also calls himself 'brother of the 
(great) king,' but which king? Since he also struck coins with the legend in Greek 
'great king Spalirises' on the obverse and in Kharoshthi on the reverse the same in the 
name of Azes, either a family connection or a political connection of the two may be 
presumed. Then, since Azes strikes some coins with Azilises, a continuation of rule if 
not in the same family at least in the same system of succession may be assumed, which 
seems to be different from that of the Bactrian- or Indo-Greeks or of the later 
Kushans. The existence of a second Azes seems assured by the coins. On the other 
hand, when one ruler overstrikes another's coins, it usually is a sign of defeat or 
conquest of the lands of the overstruck ruler, although it may mean usurpation or a 
similar act when we find a Spalirises overstrike on a coin of Spalagadama and 
Spalyrios, and then of Azilises on coins of Spalirises and also on coins of Azes I. 66 

With a new ruler, probably first in Arachosia, called Gondophares, we find some 
hints of his rule in sources other than coins, but great controversy has raged over his 
identification with the king Gaspar (with many variants) in Christian tradition. 67 

62 Dobbins, Coinage and Epigraphy [n. 55], 53- 65 Cf. Frye, "Remarks on Kingship in Ancient 
54, while Herzfeld, op. cit. [n. 28], 96, identifies Iran," AAH, 25 (1977), 78-82, with further 
Vonones with the Parthian great king of the same references. 

name, and calls him a member of the Suren family. 66 On overstrikes conveniently see Dobbins, 

Joint coins of Azes and Vonones, contrary to Coinage and Epigraphy [n. 55], 169-71, as well as 

Herzfeld, apparently do not exist. other numismatic literature of Simonetta. 

63 This is not the place to discuss theories of 67 Discussed with references in Van Lohuizen- 
matrilineal as opposed to patrilineal descent and de Leeuw, op. cit. [n. 56], 353-55, also by Konow, 
whether they played a role in the practice under op. cit. [n. 53], xliii. The name is attested elsewhere, 
discussion, as Herzfeld, op. cit. [n. 28], 94-95, as early as the Behistun inscription. It is possible 
believes, nor oftanistry as practiced by the Irish in that a certain Orthagnes was the predecessor of 
their kingship. The subject of tanistry has many Gondophares, or he may have been later. Some 
ramifications. scholars have proposed that Orthagnes is an epithet 

64 On the Parthians, see Altheim, op. cit. [ch. 6, of Gondophares since the name appears only in 
n. 19], 621, and N. Debevoise, A Political History Greek while Gondophares appears in Kharoshthi. 
o/Parthia (Chicago, 1938), 40. 



200 Chapter VII 

This has been used to date Gondophares, but it can only be used roughly, for the 
tradition of St. Thomas itself must be later, but how much later? Following the 
principle that in such a legend as that of St. Thomas and his journey to India the 
reference to the name of Gondophares is not simply gratuitous, then no matter what 
the date of the legend, it seeks to pin the journey to a historical reality, and this fits well 
with the assumed dates of Gondophares solely on the basis of coins. From the 
distribution of find spots of coins of Gondophares, it seems that he ruled from Seistan 
to the Punjab, a large empire. He introduced a symbol or coat of arms on his coins 2 
which is found also on the coins of his successor Abdagases, who on some of his coins 
in Kharoshthi has 'son of the brother of Gadaphara,' thus presumably the nephew of 
Gondophares. The enigmatic coins of Sasa or Sasas do not have this symbol, although 
the Kharoshthi legend on some of his coins mentions his father's name Aspabhata, and 
on others we find Gudaphara, presumably the same name as Gondophares, but with 
no relationship implied. 68 The symbol appears again much later as the coat of arms of 
Shapur I the Sasanian where it is represented on his horse-covering on the relief of 
Ardashir at Firuzabad, and it is impossible to discover whether the symbol is a family 
sign (of the Suren family) to which the Sasanians might have been attached, or there is 
some other connection between Gondophares and the Sasanians. In any case, it does 
not appear in the east again. 69 

Other coins are not easily arranged in a sequence. For example, we have coins of 
Indravarma, who calls himself son of Vijayamitra, followed by his son Aspavarma, 
who bears the Greek title strategos, but the Greek legend has 'great king of kings Azes,' 
probably an Azes II. 70 Since their coins are found in northwest India, mostly from the 
Taxila mint, they were probably governors of, and after, Azes II. 

Since coins of Gondophares were copied after his rule, the existence of a second 
Gondophares may be postulated. Even more enigmatic are those who struck 
imitations of the coins of Hermaeus, one of the last Greek kings of the Paropamisos, 
for it would seem that these chiefs or kings, perhaps Sakas, continued to strike 
imitations of Hermaeus down to the expansion of the Kushans in the first century of 
our era. It is clearly impossible now merely on the basis of coins to determine which 
Saka or Indo-Parthian ruler seized which area from the other, or from a Greek ruler, 
and how long it was held, although numismatists have endeavored to do this on the 
basis of monograms and find spots. With the discovery of more hoards and more 
excavations, however, the statistical probabilities would rise for identifications. 

Some coin issues are especially illusive, including two series of a certain Arsaces, or 
perhaps two rulers with the usual name for Parthian kings on their coins. Another is 
Pakores, both series of which ruler cannot be simply western Parthian issues, 

Jenkins/Narain, supra, Coin-types, 18-20. or possibly that Gondophares, or a successor. 

Many scholars have assumed that the name, written countermarked various coins which were current, 

as Sasasa, must be Sasan, but Scythian names Sasa Cf. E.J. Rapson in the Cambridge History of India, 1 

or Sasas from South Russia exist; cf. Zgusta, op. cit. (Cambridge, 1 922), 578. 

[n.53], 142. Dobbins, op. cit.[n. 55], 124, identifies 70 Jenkins/Narain, supra. Coin-types, 23. The 

Aspabhata as Aspavarma, which is unproven. relation of this Azes II with Gondophares is 

The Gondophares symbol is found counter- unknown, but maybe there is some connection 

struck on some coins of the Parthian king Orodes I through the satraps Indravarma and Aspavarma, 

(c. 90-77 B.C.) and Artabanus HI (c. a.d. 12-38), although they may simply be satraps, first under 

which indicates either a family usage (as the Suren) Azes II and then under Gondophares. 



Greco-Baclrians, Sakas and Parthians 201 

however, since they have Kharoshthi legends on the reverses. Since Pakores 
overstruck a coin of an early Kushan ruler Soter Megas, he should be placed late in the 
first century A.D. 71 Likewise the coins of Sanabares, with Parthian letters on some of 
his coins, present many problems, but since his coins have been found far to the north 
in Merv and Herat, we may regard him as a Parthian sub-king or independent ruler, 
and he should not be considered as belonging to the group of Gondophares. 72 This by 
no means exhausts the names of rulers or coins. 

To summarize the period from the fall of the Greek kingdoms north of the 
Hindukush, one must at the outset say that conjectures abound, and one can only hope 
to present a picture which is generally correct, although the details will change with 
new discoveries, mainly inscriptions or coins. The Greeks, although they relinquished 
their rule of domains to the north of the Hindukush to nomadic invaders, did so 
probably not in a swift but rather a prolonged process. Greek kings continued to rule 
in the mountains and in northwest India after 130 B.C., when Sakas came there from 
Seistan and Arachosia, although some may have come from the north, from Sinkiang. 
The accession date of Maues, apparently the first Saka king who issued coins in the 
northwest of the sub-continent, is unknown but 90-80 B.C. is a likely surmise. 
Somewhat later, about 80—70 B.C. to use decades, Vonones ruled as an independent 
Parthian king in Seistan and Arachosia. Since coins of Maues are not found in the last 
two areas, we may regard the two rulers as contemporaries, though possibly only 
overlapping a few years in their reigns. How to crowd the many Greek kings into 
northwest India, Gandhara and the Paropamisos before this time is a difficult 
problem. From the number and find spots of their coins, Apollodotus, Archebius and 
Hermaeus were the most important of the last Greek rulers. The last named had his 
coins copied extensively, and one may postulate rulers of a local kingdom or 
kingdoms in the mountains who continued the traditions of this final Greek king in 
their area, sometimes probably acknowledging the overlordship of a greater Saka or 
Parthian king who, though he issued his own coins, may have allowed the local 
population to continue with traditional coinage. There may have been Saka chiefs 
who issued the pseudo-Hermaeus coins, as Dobbins suggests, but simply the local 
population might be better candidates. 73 There also may have been a revival of Greek 
power in some limited areas of eastern Afghanistan and northwest India with names 
on coins such as Telephus, Hippostratus, Strato, and of course Apollodotus II. Further 
Greek rule in the Punjab also has been assumed. The Sakas, having ended Greek rule, 
in turn were dominated by Parthian rulers beginning with Gondophares in the first 
part of our era. Since the Parthians were mixed with Sakas, they probably provided a 
ruling class, if not a ruling bureaucracy too, for the area extending from Seistan up to 
the Punjab. From the north, the Kushans, who had consolidated their power under a 
series of capable rulers, became the rivals of the Parthians and eventually swept 
various rulers aside to form a large empire. The legacy of the Greeks was carried by 
the Kushans perhaps more than by the Sakas, who had already influenced the Kushans 

71 A. Simonetta, "An Essay on the so-called have the Gondophares symbol but one similar 
*Indo-Greek' Coinage," EW, 8 (1957), 53, plate 3. though not to be confused with it. The various 

72 Cf. K. W. Dobbins, "Sanabares and the reconstructions of numismatists are intriguing but 
Gondophares Dynasty," NC, series 1 1 (1971), 140. frequently go far beyond the evidence. 

The Sanabares coin, contrary to Dobbins, does not 73 Dobbins, op. cit. [n. 55], 128 et passim. 



202 Chapter VII 

with their Iranian 'steppe culture.' The Parthians, to the west, at this time were weak 
and could neither oppose the Kushans nor help their brethren in Arachosia and in 
India, and we may suspect that the eastern part of imperial Parthian domains, such as 
Seistan and Merv, were independent under Sanabares and his successors. We do not 
know what happened in Arachosia after the first century A.D., but Kushan authority 
seems to have extended over this province, if one judges by the number of Kushan 
coins found there. 74 The debasement of silver coinage after the reign of Azilises may 
reflect an economic decline in this part of the world. The economic importance of 
Seistan, Arachosia or the Kabul area in comparison with Gandhara and the Punjab 
should not be overestimated, but after Azilises, debased coinage was the rule until the 
imperial Kushans installed a gold standard. Seistan probably remained independent 
down to the Sasanian conquest, if we consider the coins of Arda Mitra to be 
prototypes of early Sasanian coinage. This will be discussed in the chapter on the 
Sasanians under the early eastern expansion of the Sasanians. In any case, the Kushans 
became the heirs of the Greeks and Sakas in the east, as the Sasanians did of the 
Parthians in the west. 

One must not forget that this period of the history of eastern Iran, Afghanistan and 
northwest India, as in Iran proper, was one of independent 'sub-kings' and satraps, a 
'feudal' society with much warfare and lack of central control. This is in contrast to 
the rise of Rome and the formation of the Roman Empire, but as frequently in 
history, this lack of centralized rule should not be interpreted as a period of cultural 
decay and unproductivity, for the material remains of the Sakas are impressive, 
especially the tombs of the Saka princes, such as the six kurgans of Tilla Tepe in 
northern Afghanistan where over 20,000 pieces of gold objects, such as buttons, 
pendants, plaques and weapons have been found. Although these tombs may belong 
to early Kushan princes rather than Sakas, it is impossible to differentiate the two. 75 In 
the tombs gold coins of the emperor Tiberius (a.d. 14-37) and a silver coin of 
Mithradates II of Parthia have been found, but so far no Greco- Bactrian or Kushan 
coins. The artistic finds, however, reveal widespread connections with Chinese, 
Indian, Greek, Saka (or 'Steppe') art, and the local art productions of Bactria, all 
evidence of a wealthy and ostentatious society. The art of these nomads, however, in 
spite of its striking but eclectic nature, was not the art of a mighty, centralized empire 
as that of the Achaemenids. Not only was the architectural grandeur of the 
Achaemenids missing, but the variety rather than the uniformity of smaller arts 
emphasizes the contrast with the imperial art of their predecessors. Of course, after the 
Achaemenids the entire east had been conquered by Hellenism, which was similar to 
modern 'Westernization,' but as today so then it was more or less limited to the upper 
classes. Some have equated the nomadic invasions with a revenge or a barbaric 
conquest of Hellenism by old Iranian culture, but if accepted, this could only be true 

74 D. W. MacDowall and M Ibrahim, "Pre- 75 Cf. V. I Sannidi, The Treasures of Golden 

Islamic Coins in the Kandahar Museum," Afghan Hill," ASA, 84 (1980). 125-31 . Greek inscriptions 

Studies, 1 (1978), 71-73. Overstrikes of coins of on a ring and gold bowl, ivory combs from India, 

Gondophares on those of pseudo-Hermaeus, and Chinese mirrors, and Siberian or 'steppe' gold 

by Kujula Kadphises on coins of Gondophares plaques, all remind one of later finds from Begram. 
indicate a close temporal span as well as geographi- 
cal contiguity of all of them 



Greco-Bactrians, Sakas and Parthians 203 

in western Iran, or more particularly in Persis where Achaemenid traditions or 
memories were preserved. In the east the nomadic rulers adopted and adapted 
Hellenistic culture in its manifold aspects, and just as much later after the Arab 
conquests, so too after the nomadic conquests, including the Parthians, a new 
oecumenical culture was created in the east, primarily a Greco-Iranian syncretic 
culture with the three elements of Greece, nomadic Iran, and the ancient Near Eastern 
legacy of the Achaemenids fused. It was the merit of D. Schlumberger to have 
emphasized this syncretic art in his writings. 76 Before him the attention of art 
historians was riveted to the prolific production of Gandharan art, for other than 
coins it was the Buddhist art of northwest India which gave evidence of Hellenism in 
the east. 

The problem, as seen in the past, was how to join the marvelous Greek coins of 
Bactria, which vanished after 1 30 B.C., in the minds of the investigators, to the profuse 
manifestations of western Classical influence in that Buddhist art of the Kushans and 
after, mostly dating after the second century of our era. 77 There was a hiatus of several 
centuries, and Alfred Foucher tried to bridge this gap in excavations in Balkh, while 
others tried elsewhere, yet the gap remained, such that Foucher despaired and began 
to speak of the 'Bactrian mirage' of Hellenism in the east. This led others to postulate a 
new Roman impulse on the Kushans and the designation of Gandharan art as a 
Roman provincial art, and the nomads were forgotten, since they obviously had little 
to contribute. The art of Gandhara will be discussed under the Kushans, but here 
suffice it to note that excavations at Ay Khanum have revealed a flourishing Greek 
culture and art in Bactria, while Surkh Kotal has given us a Greco-Iranian syncretic 
art, the early Kushan syncretism without Buddhism and Indian influences of later 
periods. So a link has been found in the culture of the nomads, primarily the Sakas, 
and its fusion with the settled culture which the nomads found in Bactria and 
elsewhere. 

One tends to view history in terms of power which then is equated with 
civilization. The Achaemenid Empire, Alexander, then the Seleucids, the Roman 
Empire, the Sasanians, and the comparatively recently discovered Kushans in the east 
are the foci of historical surveys, and other periods are considered barbaric, periods of 
decline or inactivity. This view, of course, has much to do with the absence of sources 
or imposing monuments left to posterity, but it neglects the changes among the 
populace in favor of stability and the ability of centralized governments to control 
and organize masses of people to build a pyramid, a Persepolis or a Roman forum. Just 
to mention art, it is questionable whether Roman copying of Greek art in the service 
of the empire was more true to the genius or spirit of Greek art than the popular (or 
non-imperial) adaptations of it in the east. The syncretic Greco-Iranian art of the Sakas 
and other nomads, although frequently depicted as stereotyped, can be characterized 
as more dynamic than the slavish copying of Greek prototypes by imperial Romans. 
There is no evidence, but one may suggest that the tribal councils of nomads had more 
in common with the agora of a Greek polis in Bactria than with the ancient Near 

76 In several articles, but see especially his 77 A popular but interesting survey of ideas 

L'Oriem Hellenist, (Paris, 1969), introduction and about Gandharan art is given by Wheeler, op. cit. 
40 -58. [n. 37], 149-71. 



204 Chapter VII 

Eastern theories of monarchy, or even with Roman imperial institutions (not 
republican). Since our sources are so meagre there is no documentation for this other 
than the evidence of material culture, but it is certainly worth considering the possible 
role of east Iranian or Central Asian influences on the history of the entire Near East, 
and not just western Iran. The role was important not only in the syncretic Greco- 
Iranian art of the west, but also in ideas from this part of the world which came to the 
west through the intermediary of the Parthians, to whom we now turn. 



CHAPTER VIII 
THE PARTHIANS ON THE PLATEAU 



Literature: The relatively recent but now classical work by N. Debevoise, A Political History o/Parthia 
(Chicago, 1938) has not been superseded, for newer general books such as M. A. R. Colledge, The 
Parthians (London, 1967) or his Parthian Art (Ithaca, N.Y., 1977), with a large bibliography, or G. A. 
Koshelenko, Kultura Parfii (Moscow, 1966), add little to the work of Debevoise, except in the realm of art 
and culture. Monographs and articles on special subjects, especially Roman-Parthian relations, however, 
have added to our knowledge of the foreign affairs of the Parthians, while the results of archaeological 
excavations, as usual, provide new sources to augment our picture of Iran under the Parthians. It must be 
noted, however, that most of the excavations are not in Iran but outside the geographical boundaries of 
the present country. The sites and materials are as follows: 

1. Nisa, near Ashkabad in Turkmenistan, was excavated by the Southern-Turkmen Combined 
Archaeological Expedition from 1948 to 1961, and many Parthian ostraca as well as material remains 
were uncovered. For a bibliography on the finds at Nisa see I. M. Diakonoffand V. A. Livshits, Parthian 
Economic Documents from Nisa, CII, Plates I — III, and Texts I (London, 1976-80), and M. E. Masson, 
Perechen opublikouannykh rabot i materialov po tematike YuTAKE (Ashkabad, 1970), a bibliography of 
over 500 items. 

2. Kuh-e Khwaja in Seistan. For a bibliography on the surveys of M. A. Stein and E. Herzfeld see L. 
Vanden Berghe, Bibliographie analytique de Varchiologie ie I'Iran ancien (Leiden, 1979), 28-29. The 
fragmentary wall paintings found here are important for Parthian art, but they have been lost. 

3. Kangavar in Media. The work of Kambakhsh-Fard and his articles relating to it may be found in 
Vanden Berghe, op. cit., 142. The temple of Anahita is the significant Parthian survival here, but later 
Sasanian changes or additions to the site have confused the picture. 

4. Bard-e Nishandeh and Masjid-e Sulaiman in Khuzistan. The work of R. Ghirshman culminated in two 
volumes on Terrasses Sacries, in MDAFI, 45 (Paris, 1976). The two sites are remains of the kingdom of 
Elymais rather than Parthian, and the inscriptions are in a Semitic language rather than Parthian. Cf. 
Vanden Berghe, op. cit. 85-88. 

5. Shahr-e Qumis near Damghan. The Parthian capital of Hekatompylos has been surveyed with sondages 
by D. Stronach and J. Hansman. For a bibliography see Vanden Berghe, op. cit. 25-26. 

There are other minor sites in Iran, usually large sites with a Parthian stratum, and excavated or 
surveyed major towns influenced by the Parthians, or under their rule, are located outside of the 
boundaries of the present country, mostly in Iraq. Among them are Hatra, Nippur, Assur, Uruk, and in 
Syria, Dura Europos and Palmyra, on which see Vanden Berghe, op. cit. 259-60. 

For Soviet works on the Parthians see T. N. Zadneprovskaya, "Bibliographie de travaux sovietiques sur 
les Parthes," SI, 4 (1975), 243-60. The works of Koshelenko are especially noteworthy, since he is the 
foremost Soviet specialist on the Parthians See also the bibliography on archaeology in succeeding issues 
of AMI. 

As with the Greco-Bactrians and Sakas, numismatics is of paramount importance in establishing the 
order of the Parthian kings, but it is more complicated than in the east because the quasi-title 'Arsaces' is 
used on most coins and not the personal name of the rulers. Other than Vanden Berghe, op. cit., 262-68, see 
A. M. Simonetta and D. G. Sellwood, "Again on the Parthian Coinage from Mithradates II to Orodes II," 
Quaderni ticinesi di numismatica e antichita classiche (Lugano, 1978), 95-119, and bibliography. 

The articles on the beginnings of Parthian history, frequently repetitions, by J. Wolski, are too many to 
list, but see one of the latest, with bibliography of others, "L'origine de la relation d'Arrien sur la paire des 
freres Arsacides, Arsaces et Tiridate," in Studies in the Sources on the History of Pre-Islamic Central Asia, ed. 
by J. Harmatta (Budapest, 1979), 67-74. The booklet by B. P. Lozinski, The Original Homeland of the 
Parthians (Mouton, The Hague, 1959), is unfortunately unusable. 

Parthian inscriptions have not been assembled in a corpus, but a bibliography for them may be found in 
P. Gignoux, Glossaire des Inscriptions Pehlevies et Parthes (London, 1972), 43-44, to which may be added 
Das Parthische Felsrelief, Sarpol-i Zohab, Iranische Denkmaler, Lieferung 7 (Berlin, 1976), 16. 



206 Chapter VIII 

The prime literary sources for the Parthians are chapters 41 and 42 of Justin's epitome of Pompeius 
Trogus, Apollodorus of Artemita (most as found in Strabo) and Arrian's Parlhika, both in Frg. Hist. 156 
and 779 respectively, as well as the Parthian Stations of Isidore. The many Classical sources on Roman- 
Parthian military and diplomatic relations are listed by Debevoise, to which work the reader is referred. 



EXPANSION OF THE KINGDOM 

Information on the origins of the Parthians comes from Justin (XLI, 1) who says they 
were originally exiles from Scythia, and Strabo (XI, 515) who says Arsaces was a 
Scythian man with the Aparni, a part of the Dahi, nomads who lived along the Ochus 
(lower Oxus) River, who invaded Parthia and conquered it. He continues that some 
say he was a Scythian while others claimed he was a Bactrian who fled from Diodotus 
and raised a revolt in Parthia. Arrian (and his successor the Byzantine writer 
Syncellus) tell a story about two brothers, Arsaces and Tiridates, who were insulted 
by the (Seleucid) satrap of Parthia. So they plotted with five companions and 
overthrew him thus freeing the Parthians. 1 The parallel of Arrian's account with the 
story of the plot of Darius against Gaumata is evident, but whether the whole story 
really has a mythological basis relating to the Discorides, or heavenly twins, as Wolski 
suggests, is uncertain. 2 The details of the lives of the early kings are clouded, but first 
we should ask if the story of the (A)parni invasion is believable and if there is any 
reason for their migration southwards into the Seleucid domains in the third century 

B.C. 

The reality of the (A)parni is indicated not only by the mention of them in Strabo, 
in Ptolemy and in Justin (in the form Sparni) but also by the Middle Persian text 
called the Bundahishn, which says that one of the offspring of Sam 'gave the 
governorship of Aparshahr to Aparnak. Aparshahr is thus named because it is the land 
of Aparnak,' 3 Furthermore, Henning sought to trace east Iranian 'Parni' words in 
Armenian borrowings from Parthian, a west Iranian tongue, as well as elsewhere. 4 
Thus we may accept the migration, more likely than an invasion, of the Parni south 
from the area of Khwarazm into Parthia in the first part of the third century B.C. The 
explanation of the name as 'mountain dwellers' and then their identification as 
inhabitants of the 'upper lands' (satrapies), later Aparshahr, is hardly correct though 
ingenious. 5 Reasons for the migrations of nomads could be many, drought, a search 
for better pasture lands, or political pressure. It is possible, as suggested by F. Koske, 
either that northern Parthia, from the Caspian Sea through present Turkmenistan to 

1 Frg. Hist., Arrianos 156, 858-59. The vari- 4 Henning, "Mitteliranisch" [ch. 6, n. 57], 93-94. 
ations in the name of the satrap have been discussed 5 By Eilers, "Demawend" [ch. 1, n. 19], 347, 
frequently by Wolksi. 373, n. 244. Abarshahr, later the Nishapur area, 

2 Wolski, supra, I'origine, 71-73. He is con vine- can be well derived from 'the upper lands,' but the 
ing in his rejection of Arrian as a reliable source for Aparni are another problem, and they surely are 
the origin of the Arsacids. not attested as 'mountaineers.' The identification of 

3 Ch. XXXV, 43^14, in the edition of B. T. the Aparni with the later Avars, proposed by W. 
Anklesaria, Zand Akaslh (Bombay, 1956), 299, Haussig, Theophylakts Exkurs liber die skythi- 
where the translation is wrong, the correct word schen Vdlker," Byzantion, 23 (1953), 329, is not 
being Aparnak, our Aparni. acceptable either historically or linguistically. 



The Parthians on the Plateau 207 

Merv, was never held by the Seleucids, or that under Antiochus I short-lived forts to 
control the area were abandoned, and the whole territory was soon independent. 6 It is 
difficult to determine how much territory the Seleucid satrapy of Parthia did include, 
but in any case, it is reasonable to suppose that the northern desert areas and 
Khwarazm were not controlled by the Seleucids but did maintain an independent 
existence. 

From Soviet archaeological excavations we know that not only the area of 
Khwarazm, south of the Aral Sea, but also the land to the east on the Jaxartes River 
contained towns and settlements as well as nomads in the period after Alexander's 
expedition. The names of various tribes which survive in Classical sources have been a 
great source of confusion and dispute among scholars, especially etymologies of names 
such as Massagetai, Derbikes, Apasiakai, Sacaraucae and others. 7 It is not possible, in 
the present state of our knowledge, to assign to a Khwarazmian state of this period the 
Apasiakai, as S. Tolstov does in various publications. 8 Also the identification of the 
state of K'ang-chu in Chinese sources, in this period located in the Talas and lower 
Jaxartes region, with either the Sacaraucae, or other peoples, is hardly possible with 
our lack of written sources. It is also not possible to assert that the expansion of the 
K'ang-chtl drove the Parni south, and since information about the K'ang-chti is so 
little and shadowy, a proper history of this part of the world cannot be 
reconstructed. 9 Obviously the Jaxartes basin and other areas of western Turkestan 
were not just barren stretches of desert with a few nomads roaming over them, and 
settlements of Iranian speakers existed there, but we know little about them, and they 
seem to have had little influence on the movement of the Parni to the south into the 
province of Parthia (Khurasan). 

Here we encounter the enigmatic Andragoras, who Justin (XLI, 4) says was the 
governor of Parthia, and he was overthrown and killed by Arsaces. The literature on 
the subject has been collected by Wolski, who correctly points out that the name is 
Greek, and there is no evidence that it is a translation of an Iranian name. He further 
suggests that his unique gold coin is a commemorative issue either at the time he 
declared his independence from the Seleucids, or simply a later emission by the 



6 F. Ya. Koske, "Plemena severnoi Parfii v borbe 8 S. P. Tolstov, Drevnii Khorezm (Moscow, 
s makedonskim zavoyevaniem," VDI, part 1 1948), 244, and in English with amplifications, 
(1962), 113-25. He dwells mostly on Alexander's "Scythians of the Aral Sea area and Khorezm," 
campaigns in Central Asia, and archaeological XXV International Congress of Orientalists, Trudy, 3 
evidence, but the assertion that the entire area was (Moscow, 1963), 157-63. Whether the Apasiakai 
not under Seleucid rule may be too sweeping a are to be identified with the Pasiani as Tolstov, 
generalization. 1 62, followed by Daffina, supra, Immigrazione, 57, 

7 A summary of some of the etymologies may assert, is also uncertain. 

be found in Daffina, supra, V Immigrazione, 54-60. 9 The identification of the K'ang-chU with 

Bailey, among others, in many articles, has Kaxoyai Ekv&cu on a map (but not in the text) 

connected saka + rauca with Chinese Sai-wang of Ptolemy , or with Kangha in the Avesta or later 

'Royal Scythians,' the Mas-Sak (Massagetai) with Kang-diz is hypothetical. See Markwart, Wehrot 

'great Saka'; cf. his latest account in "North-Iranian mid Arang [ch. 3, n. 41 ], 1 88, and B. A. Litvinskii, 

Problems," BSO/1S, 42(1979), 207. Etymologies "Das K'ang-chii-Sarmatische Farn," CAJ, 16 

are always elusive and only concern us when they (1972), 250-52, with further references and 

cause a revision of history or relate to other bibliography, 
matters. 



208 Chapter VIII 

Parthians to show their connection with the Seleucids and Alexander the Great. 10 
The authenticity of the coins being subject to doubt, their minting as com- 
memorative pieces at the time of Andragoras seems more reasonable than the 
postulation of a later propaganda reason. More than the problem of Andragoras, 
however, the chronology of the early Parthians has produced a controversial 
literature. Wolski has marshalled many arguments to show that the Parthian revolt 
occurred in 238 B.C. in the reign of Seleucus II, but neither he nor anyone else has 
satisfactorily explained the 'later' adoption of an Arsacid era beginning in 247 B.C. 
rather than in 238. ' ' Yet Wolski presents the following chronology: first invasion of 
the Parni into Iran c. 280 B.C.; the Parni under Arsaces conquer Astauene (hodie 
Quchan) c. 250; crowning of Arsaces in the capital of Astauene, Asaak (or Arshak), 
according to Isidore of Charax (11) in 247 B.C.; revolt of Andragoras, satrap of 
Parthia, 245 B.C.; Diodotus proclaims his independence in Bactria, 239 B.C.; death of 
Andragoras and the taking of power in Parthia of Arsaces 238 B.C. ; conquest of 
Hyrcania by Arsaces in 235 B.C. 1 2 The first date was found by Wolski to be the time 
when Merv and Herat were devastated by nomads at the end of the reign of Seleucus 
I, who sent a general Demodamas to punish them. The nomads were the Parni, a 
division of the Dahi who at this time were moving from the north towards the 
Caspian Sea. 13 The second date is a plausible guess, for the process of expansion of the 
Parni must have taken time. The first Arsaces deserved to have his name honored, 14 as 
later Caesar and Augustus in the Roman Empire, for he probably transformed the 
marauding bands into a kingdom. Controversy exists, however, about the succession 
of early Parthian kings. 

Fortunately, new, contemporary sources, the Parthian ostraca of Nisa, have been 
added to the Classical sources. On one ostracon we find in the year 157, \ik MLK* 
BRY BR[Y ZY pryp]tk BRY '#Y BR[Y ZY] YJfeor'91 B.C. King Arsaces, grandson 



10 J. Wolski, "Andragoras etait-il Iramen ou Hyrcania, but the rest of them extend to the land 
Grec?" SI, 4 (1975), 166-69. The name is attested parallel to Aria (Herat). He continues that they 
in Greek papyri from Ptolemaic Egypt, and one overran Hyrcania, Nisaia and the plains of the 
can think of no reason why an Iranian would seek Parthians, implying raids. Although Nisaia is the 
to translate his name into Greek. Wolski convinc- land on the northern edge of the Kopet Dagh 
ingly identifies the Andragoras of this period with range, it may have been considered part of the 
the name of a high official in a Greek inscription richer valley of Astauene to the south, where 
from Gurgan under Antiochus I, and he convinc- Arsaces was probably crowned, in a town named 
ingly rejects the historicity of the Andragoras after him, as Wolski proposes. Strabo (XI, 511), 
mentioned by Justin (XII, 4) as satrap of Parthia however, says that the Aparni made war, then 
under Alexander. The coins of Andragoras present peace and war again with the settled people, a more 
similar problems to those of Sophytes and both likely course of events than any planned conquest. 
may be contemporary satraps of c. 250 B.C. This account is repeated with additions in his 

11 Cf. J. Wolski, "L'Historicite d'Arsace I er ," "Arsace I", fondateur de letat Parthe" [n. 1 1 ], 1 59- 
Historia, 8 (1959), 235, and his "Arsace Ier 99. 

fondateur de letat Parthe," AI, 3 (1974), 197. 14 The older form of the name 'Arsaces,' Arsu 

'• Wolski, "Der Zusammenbruch der Seleuki- for Artaxerxes II has been found in Akkadian; cf. 

dcnherrschaft in Iran im 3. Jh. v. Chr.," in Altheim, A. Sachs, "Achaemenid Royal Names in Babylo- 

op. cit [ch. 7, n 1], 253-54. man Astronomical Texts," AJAH, 4 (1979), 133- 

13 Ibid., 205-07. Strabo (XI, 511) says that the 35. 



Aparni, part of the Dahi, are situated closest to 



The Parthians on the Plateau 



209 



of Phriapatius, son of the nephew of Arsaces.' 15 The editors of the ostraca (Note 15, 
pp. 20—21) reconstructed the early genealogy of the Arsacids as follows: 16 



Arsaces 



Tiridates 

I 
Artabanus 



Phriapatius 
Mithradates - Arsaces (Gotarzes) 
but Koshelenko proposed a new theory of descent: 





Arsaces 1 

1 


(Tiridates not king) 

1 






1 

Arsaces 11 


l 

x (not king) 

1 

Arsaces III (Phriapatius) 

1 




1 

Arsaces IV 


1 

Arsaces V 


Arsaces VII 


1 

X 


(Phraates I) 


(Mithradates I) 

1 


(Artabanus I) 

1 


1 
Arsaces IX 




Arsaces VI 


Arsaces VIII 






(Phraates II) 


(Mithradates II) 





The name Phriapatius may appear in another ostracon, but the inscription is damaged 
and incomplete and seems to say only that an Arsaces in the year 180 (of the Arsacid 
era = 68-67 B.C.) was a descendant of Phriapatius. 17 The latter was obviously an 
important king in the dynasty, and there is every reason to identify him with the 
third king of the Parthians, but why was he more important apparently, in the ostraca 
inscriptions, than his predecessors? The proposal of Koshelenko answers this question, 
and his suggestion in regard to the Tiridates problem does attempt to reconcile Justin 
(Trogus) with Arrian (and Syncellus), and as such is a good guess. Otherwise we must 
believe either Arrian or Justin, since they conflict. The name Artabanus for the second 



15 I. M. Dyakonov and V. A. Livshits, Doku- 
menty iz Nisy Iv. do n. e. (Moscow, 1960), 113 and 
plate. The brackets are smudges on the ostracon 
and within them are reconstructions. This is a 
strange ostracon with only these two lines, whereas 
other ostraca are accounts of quantities of wine 
from vineyards in various estates. The expression 
BRY 'liY BRY, the Parthian form of these 
Aramaic masks being unknown, is attested only 
here. 

16 See the remarks of G. A. Koshelenko, 
"Genealogiya Pervykh Arshakidov," in B. G. 
Gafurov, ed., Istoriya i kuhura narodov Srednei Azit 



(Moscow, 1976), 34. This follows Justin (XLI, 5) 
who says that Arsaces II was also called Arsaces and 
was followed by Priapatius. A bibliography on this 
ostracon is given on pp. 36—37. 

17 I. M. Dyakonov and V. A. Livshits, "Novye 
nakhodki dokumentov v Staroi Nise," Prednea- 
zialskii Sbomik, 2 (Moscow, 1966), 143, n. 28, and 
plates 10 and 10a. The historical interpretation of 
these documents may not be as much as the editors 
suggested, especially in regard to the genealogy of 
the Arsacids. Altheim, op. cit. [ch. 6, n. 19], 446, 
gives Arsaces I two sons, one Arsaces II and the 
other the father of Phriapatius. 



210 Chapter VIII 

or third ruler of the Arsacid dynasty is not attested, and if one follows Justin, the son of 
Arsaces I could be assigned that name. With Phriapatius, however, we are on firmer 
ground, and one may assign his rule to the aftermath of the invasion of Antiochus III 
from whom Arsaces (II according to Justin, III according to Arrian) fled. After the 
retreat of Antiochus III, the Parthians seem to have turned their attention to the east 
against the Greco-Bactrians. Justin (XLI, 5), however, ascribes the conquest of the 
Mardi, on the southeastern shores of the Caspian Sea, to the short reign of Phraates, 
elder brother of Mithradates I, and this would be an expected expansion of Parthian 
authority at this time. We have already seen how several provinces were wrested 
from the Greco-Bactrians, but under Mithradates I Parthian expansion to the east is 
unclear. 

In Media the situation is also unclear, even more so after the discovery of a Greek 
inscription on a relief of Herakles near Behistun, speaking of Cleomenes a satrap of the 
'upper provinces' in 149-48 B.C. 18 The easiest explanation of this is to assume that the 
Seleucid satrap of Media, Timarkhos, lost some territory in the area of present Tehran 
(Rhages) and farther east to Mithradates, while the Seleucid reconquest did not 
recover this land but maintained rule in Ecbatana. 19 According to Justin (XLI, 6) 
Mithradates had to fight many times in Media, and it is conceivable that the Seleucids 
held only the city and the lowlands, to the west of Ecbatana, at the time the relief of 
Herakles was carved. 20 Since the earliest Babylonian documents dated in the name of 
Arsaces begin in 141 B.C. after the defeat of the Seleucids in Mesopotamia, we may 
assume an interval of time of several years for the Parthians to move from Media to 
Mesopotamia and to the capture of Seleucia. The defeat of Demetrius II the Seleucid 
ruler was not accomplished in one battle or the conquest of Mesopotamia in one 
campaign, and the advance of Mithradates apparently was not swift. 2 ' It generally has 
been supposed that the Parthian king was called away to Hyrcania, as Justin (XLI, 6) 
tells us, after which Demetrius II was captured by a general of Mithradates and sent in 
captivity to Mithradates about a year later in 140-139 B.C., but the Parthian king only 
lived another year himself after having secured the submission of Elymais. 22 For 
further information we must turn to the numismatists, some of whom assert that only 
under Mithradates I did the Parthian rulers begin to strike coins, and all earlier 
attributions are false. 

The problem with early Parthian numismatics is that, unlike the coins of Bactria 
and northwest India, only the name or title Arsaces appears on the coins and identities 
must be established by style, the busts, or figures and titles. Le Rider argued that since 
the Dahi, of whom the Parni were a branch, served in the army of Antiochus III, 
according to Polybius (V, 79) and Livy (37, 40), therefore the Arsacid rulers were 
vassals of the Seleucids and did not have the right to strike coins until Mithradates I 

Cf. chapter on the Seleucids, and esp 2 ' For references cf. Debevoise, op. cit. [ch. 7, n. 

Morkholm op. cit. [ch. 6, n. 93], 178-80 64], 22-23. The coinage of Seleucia indicates that 

The renaming of Rhagai-Europos to Arsakia Mithradates struck coins there the first time only 

may have occurred under Phraates I; cf. M.-L. for a year; cf. Le Rider, op. cit. [ch. 6, n. 25], 361- 

Chaumont, "Etudes d'histoire Parthe II," Syria, 50 62. 

(1973), 204. " On these events see Will, supra, [ch. 6, n. 32], 

At this time on the plateau Bacasis or Vagasis 2, 343-44, with references; also Debevoise, op. cit. 

could have served as satrap of Media under [ch. 7, n. 64], 24-25. 
Mithradates I, according to Justin. 



The Parthians on the Plateau 211 

who liberated the Parthians from this status. 23 Other numismatists support a coinage 
going back to the first king, although they admit that the number of coins greatly 
increases under Mithradates I. 24 It is difficult to support the thesis that until 
Mithradates I the Parthians were only vassals of the Seleucids and had no right to mint 
coins, while the Dahi mercenaries of Antiochus III give no indication of Parthian 
submission to the Seleucids. As we have seen in the previous chapter, both the 
Parthian king and Euthydemus of Bactria made treaties with Antiochus III, not as 
submissive vassals, but as allies. With the withdrawal of Antiochus III from the east 
and his defeat by the Romans at Magnesia in 1 89 B.C., it is difficult to believe that the 
Parthians continued to act as subjects of the Seleucids. The symbolism on early 
Parthian coins of the seated royal archer on the reverse has been explained as the 
giving of the bow, as the symbol of authority, to the king from the gods, a practice 
attested for the Sakas, and this would not suggest any position as vassals of the 
Seleucids. 25 The questions which early Parthian rulers struck coins and where the 
mint sites were located have not been answered, but inasmuch as mints do not require 
heavy equipment, but are even mobile, the location of a mint at Nisa, Asaak or Dara 
in the north Parthian homeland is not impossible. 26 

The conquests of Mithradates I brought the Parthians from a small kingdom in the 
east to a position of power in the arena of the Near East, but after the Parthian king's 
death in 138 b.c. a campaign of recovery of eastern domains by Antiochus VII began. 
Mithradates I had struck in Seleucia on the Tigris coins with the term philhellene, 
probably as a sign of conciliation with the Greek population, and possibly as a sign of a 
special relationship of that great city with the Parthian conqueror. 27 Parthian rule in 
Seleucia continued after the death of Mithradates under Phraates II until Antiochus 
VII appeared, after consolidating his position in Syria and the west. This was not 
until 131 B.C., however, as can be determined by cuneiform documents from 

23 Le Rider, op. cit. [ch. 6, n. 25], 315-22, with of this, is convincing. For an etymology of the 
references to previous scholars, such as J. de name Sinatruces see Henning, "Mitteliranisch" [ch. 
Morgan and E. T. Newell (with certain reserves). 6, n. 57], 41, n. 1. 

The attribution to the mint at Ecbatana of issues of 26 On these towns see Chaumont, op. cit. [n. 1 9], 

Mithradates I, with the title 'great king,' may be 197-222. It is hardly possible to trace a succession 

attributed to the conquest of that mint site c. 150- of Parthian capitals from Asaak, Dara, and Nisa 

147 B.C. through Hekatompylos, Arsakia-Rhages to Ecba- 

24 Cf. Wroth's catalogue of the British Museum tana as the progress of Parthian arms to the west 
and A. Simonetta, "La monetazione Partica dal 247 since the 'capital' in the period before Mithradates 
al 122 a.C." Rivista Italiana di Numismatica, 16 II changed considerably, and we are unsure which 
(1968), 20-25; M. T. Abgarians and D. G. city was a capital and which just an important 
Sellwood, "A Hoard of Early Parthian Drachms," center. 

JVC, 11 (1971), 115-18, and G. A. Koshelenko, 27 Cf. R. H. McDowell, Coins from Seleucia on 

"Neketorye voprosy rannei istorii Parfii," VDI, no. the Tigris (Ann. Arbor, 1 935), 218, and Sellwood, 

1 (1968), 53, and his "Monetnoe delo Parfii pri op. cit. [ch. 7, n. 57], 38. The copper coinage issued 

Mitridate I," Numismatika i Epigrafika, 10 (Mos- by Seleucia under Parthian rule indicates a greater 

cow, 1972), 81, hold to an earlier coinage for the autonomy for the city than it had under Seleucid 

Parthians. control; see also McDowell, Stamped and Inscribed 

25 D. S. Raevskii, "K voprosu ob obosnovanii Objects from Seleucia (Ann Arbor, 1935), 6, and Le 
tsarskoi vlasti v Parfii." in B. G. Gafurov, ed., Rider, op. cit. [ch. 6, n. 25], 373. On the coinage of 
Srednyaya Aziya v drevnosti i srednevekove (Mos- Seleucia under Mithradates see Koshelenko, op. cit. 
cow, 1977), 81-87. His analysis of the archer-chief [n. 24], 100. 

among the Scythians, and the Parthian counterpart 



212 Chapter VIII 

Mesopotamia. 28 The exact time of the march of Antiochus VII into Mesopotamia is 
difficult to establish, but if his entry into Seleucia was in 130 after winning three 
battles against Parthian forces, presumably in northern Mesopotamia, we may suggest 
that the reconquest of Mesopotamia took a number of months in 130 B.C. As the 
sources say, many local princes joined Antiochus who took the title 'great' king in 
honor of his victories. 29 The continued campaign against the Parthians on the Iranian 
plateau has presented some discord in the interpretation of meager sources. It is not 
possible to determine when precisely Demetrius II, who had remained in Parthian 
captivity so many years, was released by Phraates, while other details are also disputed. 
Coins from Susa indicate that in the year 130-129 that city reverted from Parthian to 
Seleucid allegiance and apparently many areas on the Iranian plateau also threw away 
their fealty to Phraates for Antiochus VII. 30 One extensive study of the Parthian 
campaign of Antiochus argues that he went as far as the homeland of the Parthians and 
it was there that he wintered and then lost his life in the early spring of 1 29 B.C. 3 ' This 
is unlikely, since Ecbatana is neither mentioned in any source, nor were any coins of 
Demetrius or Antiochus VII struck there, which seems odd if the latter actually 
wintered in the homeland of the Parthians. Refusing the offer of the Parthians to 
negotiate peace, Antiochus, probably in Media, was surprised with a small body of his 
troops and either was killed or committed suicide. 32 Seleucid control disintegrated 
and the Parthians re-established their rule. Phraates appointed a certain Himerus, a 
Greek to judge by his name, governor of BabyJonia and left for the east where the 
Central Asian nomads, we have seen in the last chapter, threatened Parthian rule. 33 
According to Justin (XLII, 1), Phraates had hired Saka mercenaries for his war against 
Antiochus, but the swift demise of the latter obviated any need for the Sakas who 
were not paid, and rose against Phraates. The latter moved against them with Greeks 
and others from Seleucid forces now incorporated in his own army, but these on the 
first occasion abandoned Phraates who was killed by the Sakas in 128 B.C. The nomads 
were powerful, for the uncle and successor of Phraates, Artabanus lost his life fighting 
them in 123 B.C. 34 There is no information about the state of affairs in the east after 
Artabanus, but one may presume that for a while Parthian prestige was at a low ebb. 
In the west, however, we are better informed not only because of literary notices, 



28 See references in Debevoise, op. cil. [ch. 7, n. he does not use Parthia or Parthuaia, but always a 

64], 29, n. 3. plural ii< TIap9(uv for 'from Parthia' and «s 

Ibid., 32. I7ap&ovs 'into Parthia.' Athenaeus, Deipnosophis- 

30 Le Rider, op. cil. [ch. 6, n. 25], 377-78. No tae (XII, 540) uses the more likely term 'Media' 
coins of Antiochus from the mint of Ecbatana, instead of Parthia. 

however, have been found. 32 Fischer, op. cit. [n. 31], 46, favors the suicide 

31 T. Fischer, Unlersuchungen zum Partherkrieg report. Whether Ecbatana became a Parthian 
Anliochos' VII., (Ph.D. thesis, Tubingen, 1 970), 39. capital in the time of Phraates is also uncertain. 
His insistence that in Josephus the term Parthyene 33 Debevoise, op. cil. [ch. 7, n. 64], 35, follows 
means the homeland of the Parthians while Parthia Diodorus (XXXIV, 21) and Posidonius in calling 
or Parthuaia meant the whole empire, and Himerus an Hyrcanian, which is possible, but his 
Antiochus went into the former, is questionable. name is Greek, attested in Papyrus from Egypt and 
Actually Josephus, Ant. Ind. XIII, 253, uses the he then would have been a Greek settled in 
former, on this one occasion, in an offhand remark Hyrcania or a native who took a Greek name, 
that Demetrius was released at the time that 3i There is no reason to reject the statement of 
Antiochus invaded Parthyene, whereas elsewhere Justin (XLII, 2) that he died fighting the Thogarii 



The Parthians on the Plateau 213 

but also cuneiform records and a more ample coinage aid in the reconstruction of 
history. After Antiochus VII lost his life Seleucia and much of Mesopotamia reverted 
to Parthian rule. But the weakness of the Seleucids had not only induced Bactria, 
Parthia and much of the Iranian plateau to secede from Seleucid rule, but also parts of 
Mesopotamia became independent or semi-independent. In the south the kingdoms 
of Characene and of Elymais make an appearance on the scene at this time and there is 
no reason to suppose that in the east there was more centralized control over Seistan, 
Kerman and elsewhere than in Mesopotamia. 

The Parthian state did not disintegrate, however, and the reign of Mithradates II 
(123-c. 87 B.C.) marks a high point in Parthian central power. He reconquered 
Babylonia, which had at first maintained a quasi-independent position under 
Himerus, who seems to have oppressed the local population greatly according to 
Justin (XLII, 1) and Diodorus (XXIV, 18). The former opinion, according to coins 
supposedly struck by Himerus, that he became an independent king of Babylonia, is 
no longer accepted. 35 The sequence of events is uncertain, but it would seem that 
Hyspaosines, the ruler of Characene at the mouth of the Persian Gulf, although he 
fought Himerus and others, did not occupy either Seleucia or Susa, both of which, 
according to the coinage, remained in Parthian hands. 36 Just how much Mithradates 
had to reconquer is difficult to determine, but probably not as much as hitherto 
assumed. He did, of course, defeat and secure the submission of Characene, and 
overstrikes of Mithradates on the coins of Hyspaosines exist. He took the title 'king of 
kings,' the first Arsacid to do so, and he is pictured on his later coins wearing a 
distinctive crown or tiara. Whether this indicates a new order of government of the 
Parthians with vassal states and semi-independent cities such as Seleucia and Susa is 
unknown, but at some time, perhaps under Mithradates, the Parthian state became a 
loose empire rather than a tribal kingdom, which question will be examined later. 
The long rule of Mithradates II, in any case, was a time of consolidation of Parthian 
institutions as well as expansion. For the first time we hear of Armenia, where after 
the fall of Antiochus III, according to Strabo (XI, 528) two of his generals carved out 
kingdoms for themselves, 'Greater' Armenia under Artaxias (Artas"6s in Armenian), 
and another centered on Sophene and the upper Euphrates area under Zariadres. 37 
Antiochus IV invaded 'Greater' Armenia and secured the submission of Artaxias 



(corrected to Thocarii) as Tarn, Seleucid-Parlhian Hvidberg (Copenhagen, 1953), 289-90. He well 

Studies (London, 1930), 13-14 does. This Arta- may have destroyed the city of Babylon, but 

banus is sometimes called Artabanus II, and if the Hyspaosines is mentioned as ruler there in 127- 

second ruler of the dynasty Arsaces II had the name 1 26 B.C. in a cuneiform tablet ; cf. Fischer, op. cit. [n. 

Artabanus, he was obviously the first. We do not 31], 19, for references. 

know, however, and without evidence it is better 3 * Le Rider, op. cit. [ch. 6, n. 25], 382-83, and 

to call the uncle of Phraates by the numeral one. Fischer, op. cil. [n. 31 ], 58-59, for other references. 

The passage in the prolegomenon to book 41 of See also H. Klengel, "Babylon zur Zeit der Perser, 

Tragus (ed. by O. Seel in Leipzig, 1955) has the Griechen und der Parther," Forschungen und 

following: successors (of Arsaces I) deinde eius Berichte, Staatliche Museen zu Berlin, 6 (1962), 

Artabanus et Tigranes cognomine Deus, a quo subacta 40-53. 

est Media et Mesopotamia, which is quite unreliable. 3 "' On this period of Armenian history see Ya. 

35 For his position as king see S. A. Pallis, "The A. Manadyan, Tigran Vtoroi i Rim (Erevan, 1943), 

History of Babylon 538-93 B.C.," in Studia 18-21, with references to Polybius and other 

Orientalia, loanni Pedersen septuagenario, ed. by F. sources. 



214 Chapter VIII 

which lasted only until the death of Antiochus. 38 According to Diodorus (XXXI, 
27), Artaxias became an ally of Timarkhos, the rebel Seleucid governor of Media, but 
we do not know of the rewards or land offered to Artaxias for his support, although if 
we follow Strabo the areas under the rule of Artaxias included much of present 
Iranian Azerbaijan around Lake Urmia. Artaxias was followed by his son Artavazdes, 
and it was against him that Mithradates II of Parthia led an army and secured his 
submission. The date of this expedition is unknown, but presumably it was near the 
beginning of the reign of Mithradates, for the future Tigranes the Great was taken as a 
hostage by the Parthians. According to Appian's Roman History (Syrian Wars, 48) 
Tigranes was the son of Tigranes, and the question arises whether the father ever ruled 
as Tigranes I and just what his relation was with Artavazdes. The latter may have been 
his uncle, and Tigranes I may have been put in power by Mithradates II while holding 
his son as a hostage, but these are mere surmises. 39 What seems certain is that under 
Mithradates II the first contacts between Rome and Parthia took place about 92 B.C., 
but which apparently led to no treaty between Sulla, the governor of Cilicia at that 
time, and Orobazes, ambassador of Mithradates. 40 This will be further discussed 
below. 

The last years of rule of Mithradates II are uncertain, and numismatists have argued 
about his successors, calling this the 'dark age' of Parthian history. 41 Since their many 
and varied arguments cannot be discussed here, only a summary of views will be 
attempted, with some preliminary observations on the coins, based primarily on the 
research of Sellwood and Simonetta. In the coinage of Mithradates II we find a change 
in the reverse sides from an archer sitting on the omphalos, which was probably 
inspired by the Seleucid use of the omphalos, to an archer sitting on a throne, which 
was followed by all subsequent Parthian drachms. From cuneiform tables we infer 
that between 110 and 109 B.C. the title of Mithradates II was changed from 'Arsaces, 
king' to 'Arsaces, king of kings,' and by 91 B.C. his name disappears from Babylonian 
documents. In 87 B.C. the Seleucid ruler Demetrius HI Eucaerus was sent as a prisoner 
to the Parthian king, who may have been Mithradates II. 42 In 91-90 B.C. Gotarzes is 

38 Boundary stones ofthis king have been found vol. 6 NC (1966), 15-40; K. W. Dobbins, The 
in Armenia written in Aramaic similar to Successors of Mithradates II of Parthia," series 7, 
Achaemenid chancellory practice. In them he is vol. 15 NC (1975), 19-45; D. Sellwood, "The 
called Artaxerxes son of Zariadres of the Orontid Drachms of the Parthian 'Dark Age'," JRAS 
(Ervanid) family; cf. A. Perikhanian, "Les Inscrip- (1976), 1-25; N. Waggoner, "The Coinage of 
tions arameennes du roi Artaches," REA, 8 (1971), Phraates III of Parthia," in D. K. Kouymijian, ed., 
1 69-74. Apparently he ordered a measuring of Near Eastern Numismatics, etc. in Honor of George 
land in 1 79 B.C. C. Miles (Beirut, 1 974), 1 5-26 ; and Simonetta and 

39 Marquart, ErdnSahr [ch. 3, n. 37], 173, Sellwood, supra, "Again Coinage", 119, with 
suggested that Mithradates I of Parthia fought bibliography. 

against Artavazdes, but this period of history is * 2 Sellwood, op. cit. [n. 41], 6; Simonetta and 

lacking in sources. Manadyan, op. cit. [n. 37], 26, Sellwood, supra, "Again Coinage", 104, and 

suggests that Tigranes I preceded Tigranes II, his Debevoise. op. cit. [ch. 7, n. 64], 50. The sources are 

son, as king. not certain that the ruler to whom Demetrius was 

4 On the sources, and for a good analysis see J. sent was Mithradates, in spite of Simonetta's 

Dobias, "Les premiers rapports des Romains avec assertion. Later tetradrachms, which were struck 

les Parthes et l'occupation de la Syrie," AO, 3 only at Seleucia, do give personal names and dates 

(1931), 218-21. of Parthian rulers and are thus invaluable for 

A. M. Simonetta, "Some Remarks on the identification of rulers. 
Arsacid Coinage of the Period 90-57 B.C.," series 7, 



TTie Parthians on the Plateau 215 

mentioned in cuneiform tablets as king, but he did not rule more than a year, for 
succeeding dated tablets until 87-86 B.C. mention an 'Arsaces who drove out 
Gotarzes,' who is not further identified but may be identified as Orodes I. Then a 
tablet dated 80-79 B.C. has an 'Arsaces who drove out Orodes,' who is probably the 
Arsaces called 'Philopator on his coins. Opinions based on coins alone may be widely 
divergent. For example, countermarks of a certain Otannes on coins of Sinatruces and 
Phraates III have been assigned to a satrap of the Parthians in Persis, functioning 
as an intermediary between the Persis rulers Autophradates II and Darius II, by 
Koshelenko, while Simonetta makes Otannes a last Iranian independent ruler 
preceding Gondophares, the Indo-Parthian ruler. 43 

A badly damaged relief with a Greek inscription at Behistun has been restored in 
the second line as reading 'Gotarzes, satrap of satraps, the great king Mithradates,' in 
the nominative instead of expected genitive case. 44 The first line mentions another 
name, perhaps the one who ordered the relief carved in the time of Mithradates and of 
Gotarzes, with the unique title not found elsewhere. It has been proposed that the 
Gotarzes of this inscription is the one mentioned in cuneiform tablets from 
Babylonia, which may be true, but we have no information regarding his 
relationship with Mithradates or with Orodes I, presumably his opponent. In this 
period of confusion Tigranes II of Armenia expanded his power and domains against 
the Parthians and exacted revenge for the losses suffered by Armenia. 

Tigranes II had been held as a hostage but then was placed on the throne of 'Greater' 
Armenia by Mithradates. Either after the death of Mithradates or during his last 
years, Tigranes not only took back seventy valleys which the Armenians had been 
forced to cede to Mithradates, but he also secured the submission of the rulers, or 
satraps, of Atropatene (Azerbaijan) and Gordyene (Upper Tigris), as well as parts of 
Mesopotamia, and western lands with which we are not concerned, according to 
Strabo (XI, 532). Tigranes apparently raided almost as far as Ecbatana and took the 
title 'king of kings,' which the Parthian kings in that time did not hold according to 
their coins. This state of affairs must have lasted until about 68 B.C. when Lucullus, the 
Roman general, took Nisibis, although, since Tigranes was much occupied in the 
west, he (Tigranes) may have lost territory to the Parthians earlier. The situation must 
have been confused and bad for the Parthians, such that Sinatruces or Sanatruk, 
presumably an Arsacid prince, was brought to the throne by the Sacaraucae about 77 
B.C., but it is not known whether the nomads were asked by Sinatruces to aid him or 
whether they took advantage of the chaos to push their own interests. Since he was 
reportedly eighty years old, he must have been closely related to one of the former 
Parthian kings, otherwise a younger man would have been more appropriate. He 

43 G. A. Koshelenko, "Drakhma Sinatruka s 45 The main source are the Makrobioi (Ps. 
nadchekankoi Otany," Numismalika i Epigrafika, 9 Lucian) 15; for a discussion of this passage see 
(1971), 33-37, and Simonetta/Sellwood, supra, Daffina, supra, Immigrazione, 75-76. After Sina- 
"Again Coinage", 107. Koshelenko bases his truces the significance of Parthian numismatics 
surmise on the cap or bashlyk worn by Otannes and changes from an exclusive source to one in support 
his name, which he claims is found only in Fars in of literary notices. 

ancient sources. This is most dubious. 

44 E. Herzfeld, Am Tor von Asien (Berlin, 1 920), 
35-39. 



216 Chapter VIII 

ruled about seven years, and his coins are found in Susa, as well as elsewhere, an 
indication that he recovered most of Iran, including Babylonia, for the Parthians. 
With the accession of Phraates III, son of Sinatruces, sometime between 70 and 68, 
for the actual date of death of his father is uncertain, the history of Parthia becomes 
connected with Roman history, and the sources increase considerably. Also from the 
time of Phraates, we may suggest that institutions are fixed and the next two and one 
half centuries down to the fall of the Parthians are filled primarily with internal 
struggles for supreme power, and with the external wars with the Romans. The 
eastern frontiers seem to have been more or less stable, especially with the rise of the 
Kushans and their creation of a centralized state; thus Roman-Parthian relations are 
of paramount importance. It seems appropriate that before discussing this com- 
paratively well-known phase of Parthian history we should examine the structure 
and institutions of the Parthians as far as we can reconstruct them. 



THE ORGANIZATION OF RULE 

The position of the Parthian king in the eyes of his subjects probably changed as the 
empire grew out of a small state with a background of nomadic traditions of rule. The 
ideology of kingship would be composed of three traditions, or rather an amalgam of 
the nomadic (development of Indo-Iranian concepts), Greek, and ancient Near 
Eastern (primarily Achaemenid) traditions. 46 It is, of course, difficult to disentangle 
the three, but elements of all appear in Parthian royal ideology, as much as it can be 
determined from coins (legends as well as crowns), from a few details in literary 
sources, and from our knowledge of both earlier and later practices and beliefs in Iran, 
or those of neighboring peoples. The nomadic, or what might also be called Indo- 
Iranian, beliefs probably included the concept of the royal fam or 'glory' which was so 
prominent among Iranian peoples everywhere. The belief in the investiture of the 
chief of a tribe with a bow by a deity, as pictured on the earliest coins, has been 
mentioned, and one may elucidate here briefly the idea of the royal 'glory' or 
'fortune.' Much has been written about the fam, its etymology and meaning, but here 
we are only interested in any significance it had for the Parthian rulers. 47 At the outset 
one should note the existence of names with fam in them on the ostraca from Nisa 
such as Prnhw, Prnbg and Mtrprn, literally 'the glory of Mithra.' 48 The fam is an 
ancient concept, and in the Avesta we hear of the fam or 'glory of the Aryan lands' 
(Yast, 19, 57 and 64 foil.), whereas later we learn of the fam of the 'king of kings' or 
the farr-e izadl or 'divine farri as found in the Shahname, and this fam is the guide of the 
king in his rule. 49 When Taeger says that the Parthian kings had a dynastic cult 

It is interesting to see the same tripartite "A v. xvaranah," Bulletin of the Society for Near 

division applied to architecture, based on religious Eastern Studies in Japan, 17 (1974), 75-86, English 

beliefs, by G. A. Koshelenko, Kultura Parfii summary p. 183, where the word is connected 

(Moscow, 1966), 33. with 'eating.' On the charisma of the king see F. 

The fundamental work on the word and its Taeger, Charisma, 1 (Stuttgart, 1957), 304-431. 

etymology is by H. W. Bailey, Zoroastrian Problems 48 For references see Litvinskii, op. cit. [n. 9], 2, 

in the Ninth-Century Books, 2nd ed. (Oxford, 50. 

1971), 1-78 and xvii-xl; J. Greppin, "XvarSnah," 49 For a study of the concept in the Shahname see 

JIES, 1 (1973), 232-41, with references to articles Manuchehr KhudSySr MahabbT, Farr-e yazdan dar 

byj. Duchesne-Guillemin and others. Cf.E.Imoto, ta'rlkh-e adyan (Teheran, 1972), 1-42. For a 



The Parthians on the Plateau 211 

without a noumenal or divine character, this is true if he is refuting the divinization of 
the ruler, but it is not so if one considers the^jrn as a divine concept. 50 The dynastic 
cult, or the cult of ancestors, may be characterized as the same as the Hellenistic cult of 
dead heroes, combined with an ancient and widespread cult of ancestors, to make a 
royal cult, the implementation of which was coterminous with the state's boundaries. 
The Seleucid and Greco-Bactrian kings' use of the epithet #eo? on coins apparently 
was bound to the title or office of king, or to the institution of kingship, and was not a 
personal divinization, and this was the same for the Arsacids. 5 ' The Hellenistic ruler 
cult was adopted by the Parthians, but probably not at the beginning of their rule, 
when Central Asian or tribal notions of kingship were predominant. So we may 
conjecture that Arsaces I convinced his friends that the royal jam had descended on 
him and his family, which was destined to rule an even larger area than at the 
commencement of his rule. Obviously power and success in conquest were elements 
in the rise of the Parthians. In time, Achaemenid concepts of kingship joined the 
ancient practices and beliefs together with the heritage of Hellenistic kingship of the 
Seleucids. This may be seen in the adoption of the title 'king of kings' by Mithradates 
II and by his use of a distinctive crown or tiara instead of the usual diadem of the 
Hellenistic rulers. Koshelenko has described the development of Parthian ruling 
ideology on the basis of their coins, first with the simple name Arsaces without a title, 
which he claims shows that Arsaces like Diodotus did not revolt against the Seleucids 
but simply took over the right of coinage because of the absence of Seleucid 
authority. 52 Also the existence of an ever burning fire in the town of Asaak where 
Arsaces I was first proclaimed king, as attested by Isidore of Charax 1 1, suggests that 
the act of proclaiming a king had a religious significance. One could also suggest that 
the formation of a royal ideology began with Arsaces I, who must have had a strong 
personality to have given his name as a title to all succeeding Parthian kings. This fact 
in itself attests to the importance of a royal family or royal clan of the Arsacids in the 
political ideology of the Parthians. The protocol of crowning of the king was 
undoubtedly also developed over time, usually with the head of the Suren family in 
the capacity of the one who put the crown on the king's head. 53 

discussion of the concept of Jam in Iran and H. Dorrie, Der Konigskult des Antiochos von 

neighboring concepts see F. Dvornik, Early Chris- Kominagene, Abh. GWC, 60 (1964), 236 pp. 

lian and Byzantine Political Philosophy, 1 (Cam- 52 G. A. Koshelenko, "Tsarskaya vlast i ee 

bridge, Mass., 1 966), 85. For Hellenistic influences obosnovanie v rannei Parfii," in B. G. Gafurov, ed., 

and a discussion of headgear, see Ritter, op. cit. (ch. lsloriya Iranskogo Cosudarstva i Kultury (Moscow, 

6. n. 28], 27-30, 125. 1971), 213. 

50 Taeger, op. cit., 1 [n. 47], 432. The concept of " On the crowning of various kings, with 
Jam was more than a mere charismatic exaltation sources, see Ritter, op. cit. [ch. 6, n. 28], 168. Trajan 
of the king (p. 402) but was similar to Greek tu^t; crowned his own Arsacid candidate for the 
and Sofa. Neither the Indo-European ramifica- Parthian throne in Ctesiphon, but this only indi- 
tions of the concept nor the religious implications cates the symbolic importance of the ceremony, 
of kingship can be discussed here, for which see G. For the crowns see H. von Gall, "Beobachtungen 
Widengren, "The Sacral Kingship of Iran," Numen, zum arsakidischen Diadem und zur parthischen 
Supplement 4 (Leiden, 1959), 424-55. Bildkunst," Istanbuler Mitteilungen, 19/20 (1969- 

51 Taeger, op. cit., 329, 333, 432. Herakles, a 70), 299-318, and for the curious crown of 
divinized hero, was made the ancestor of the Phraates III see J. Sugiyama, "Some Problems of 
Seleucids. As Taeger aptly says, the Hellenistic Parthian Kings' Crowns," Orient, 9 (Tokyo, 1973), 
ruler cult had many opponents but hardly any 31-41. 

martyrs (p. 307). On the Hellenistic ruler cult see 



218 Chapter VIII 

Much has been written about the institution of 'double kingship' in Iran, but it 
seems that the association of the son (crown prince) or successor of the king with the 
ruler to insure continuity has been mistaken for the institution of two kings, as in 
ancient Sparta and elsewhere. 54 In any case, there is no evidence for an institution of 
two kings among the Parthians, even if etymologically the title of bidakhsh, which we 
meet below, meant 'second king,' which itself is uncertain. 55 The formulae and 
protocols of rule were highly developed under the Parthians, but so were they under 
the Roman emperors. How much of the hyperbole of kingship, such as the comment 
of Ammianus (XXIII, 6, 6) that the Parthians worshiped Arsaces as a god, can be 
believed as real and how much of it is political propaganda is difficult to determine. 
With Mithradates II, however, we can see the evidence of such political propaganda 
on his coins and with his title. It is unlikely that his reign marks a change in political 
ideology different from the past; rather a significant progress in the continuing 
evolution or development of an ongoing Parthian royal protocol and belief is 
indicated. The usurpation of the title 'king of kings' by the Armenian Tigranes II is 
only an indication of the realities of power and the importance of the developing 
Parthian ideology in regard to kingship elsewhere in the Near East. 

The aristocracy and its assumption of titles paralleled the ideology of kingship but 
changes took place to a greater extent in this domain. Presumably the Parni, when 
they moved into Parthia, were still in a form of society characteristic of steppe tribes 
similar to the Indo-Iranian period. The division of that society into the classic 
tripartite warriors, priests and the rest of the people, we may assume, had developed 
towards a more differentiated class system of nobles of various grades, with common 
folk also divided according to professions and beliefs. Likewise we may assume a 
continuation of the heritage of the tribe, clan and family organization of pastoral 
societies. The mixture of old and new in the case of the Parthians produced a knightly 
or feudal class, each member of which was called azat, which, as Perikhanian has 
shown, should not be confused with the word meaning 'free,' although the two later 
did fall together. 56 The family of the king had a special position, especially the crown 
prince or heir apparent, the vispuhr, but as mentioned, the royal family of the Arsacids 
all participated in the heritage or patrimony of the ruler, and this special position, as 
argued by Perikhanian, was called udspuhrakanih (pp. 19-21). The proliferation and 
differentiation of titles in the Parthian court cannot be dated any more than changes 
in functions or the use of honorifics as well as titles. For example, the Parthian title 
*paiayr\w, attested in Syriac, Sogdian and at Hatra, may well have been coined by the 
Parthian kings to designate either a regent or the successor to the throne, but it may 
have been, on the other hand, an honorific designation rather than a fixed title 



54 Cf. Frye, "Remarks" [ch. 7, n. 65], 78-82, for 56 A. Perikhanian, "Notes sur le lexique iranien 
further references. et armenien," REA, 4 (1 968), 5-30. The secondary 

55 Szemerenyi "Iranica V," [ch. 4, n. 63], 391. meaning of vaspuhrakan as princes who were not 
The etymology, meaning 'second ruler,' is no sons of the king has only confused the matter. Note 
assurance of the existence of such an office; rather also that a feminine form *vi$duxta 'princess' is 
one would expect this originally to be an presumed for Parthian by Perikhanian. 
appellation given to a confidant or a close friend of 

the king regardless of office. 



The Partitions on the Plateau 219 

in a hierarchy or organization. 57 As the Parthians expanded, they encountered all 
kinds of principalities, city states, or other forms of rule which had evolved under 
the Seleucids, or which had come into existence when Seleucid rule weakened or 
collapsed. If we may take Armenia as a kind of 'microcosm' of the Parthian 
'macrocosm,' then the remark that Armenia was a collection of royal domains, 
military viceroyalties, separate principalities and temple lands may be applied with 
even more cogency to the Arsacid kingdom. 58 Furthermore, titles which appear in 
the Parthian east may not have had any currency in the western part of the territories 
ruled by the Arsacids. For example, the term kanarang or kdradranga, which we shall 
discuss in the chapter on the Kushans, is not found in the west. We may suppose that 
the Parni aristocracy intermarried with both the local, settled Parthian aristocracy, as 
well as with any Greek or Macedonian aristocracy. By the end of the second century 
B.C. they were also undoubtedly mixed with local Iranian upper classes, not to 
mention any Saka or other intruders. As Wolski has convincingly shown, after the 
consolidation of the Parthian state on the basis of tribal support of the king, in its 
expansion the ruler, following Hellenistic tradition, had to engage mercenaries to 
further his ambitions. 59 Until the first century B.C. the mixed Parthian aristocracy 
and the mercenaries fought together for common booty and for the ruler, but later 
internecine struggles between various pretenders to the throne supported by different 
groups of the aristocracy were endemic. As the boundaries of the Parthian state 
became fixed, especially in the west by the Roman Empire, revenues and booty based 
on an expansionist policy greatly declined, and the influence of the ruler fell as that of 
the aristocracy rose. For more than two centuries Parthia was on the defensive, for on 
the whole the Romans were aggressors in the wars between them. One may describe a 
change in Parthia in the first century B.C. as a transition from the old world of the 
Hellenistic monarchies to a new 'feudal' age, which is the picture of the Parthian state 
given by Classical sources. The Parthian aristocracy became wealthy and powerful as 
the influence of the king sank. At the same time the traditions, protocol and nominal 
allegiance to royalty were preserved and even fostered. 60 Society became more fixed 
in various classes and hierarchies. 

The upper nobility, many of whom were relatives of the king, were given 
provinces to govern, as were brothers or immediate members of the king's family. 61 
The relationship between these governors and a court nobility is unknown, and it is 
unwise to project backwards information from the Sasanian period except to the 

57 For references see D. Harnack apud Altheim, in Armenia parallels Parthia, but one should not 
op. cit. [ch. 6, n. 49], 516-19. Gershevitch proposed assume that the situation in Armenia was a carbon 

the form *patagrlu>, which is preferable to copy of Parthia. 
Harnacks' *paigtab. His further suggestion that the 59 J. Wolski, "Le role et Timportance des 

Parthian word is a caique on Greek SiaSoxos may mercenaires dans letat Parthe," IA, 5 (1965), 107. 
be true, but the latter was hardly a real title in 60 Cf. J. Wolski, "L'aristocratie parthe et les 

Hellenistic times. Other than the vispuhr, usually commencements du feodalisme en Iran," IA, 7 

written as BR BYT', we find 'princess' visduxta (1967), 133-44, with further references. 
(BR BYTH) and the general designations "hiynd 61 E.g., Vologeses gave the governorship of 

'prince,' iftvtwy 'lord,' as well as other words for the Media to one brother Pacorus and Armenia to 

rulers. another brother Tiridates, ace. to Tacitus, Annals 

58 R. H. Hewsen, "Introduction to Armenian (XV, 2). The Greek pcyujTa.v€S corresponds to 
Historical Geography," REA, 13 (1979), 96. Much the wuzurgttn. 



220 Chapter VIII 

period just before the end of Parthian rule. We may suppose, however, the existence 
and power of the class of great nobility, the Sasanian wuzurg&n, attached to court and 
having a high position in society because of relationship to the king and presumably 
owning extensive lands. The governors of large provinces, the hltrdrn or shahryaran, 
were the equivalent of small kings, while the majority of the nobility were small 
landowners, the azatan, or liberi of Latin sources, who brought foot soldiers with them 
from their lands when called to support the king in war. 

Much has been made of a senate or council of Parthian nobles and priests who 
supposedly elected the king from the Arsacid family in the last two centuries of 
Parthian rule. Wolski has shown that this is a mistaken assumption on the part of 
Greek and Latin authors. Rather, the nomination of the ruler at times in the last two 
centuries of Parthian rule followed a long existing institution and should not be 
attributed to a new institution, a senate which arose as a result of the weakness of the 
rulers and competing claims of various princes to the throne. 62 The privileges and the 
ranks of the aristocracy were strictly arranged, but no notitia dignitatum has survived, 
except in Armenian, for the later Arsacid court of that land. 63 It is quite possible that 
lists of military as well as civil positions existed for various provinces or lands in the 
Parthian kingdom with coats of arms for each dignity, as in the case of the late Roman 
Empire. 64 We cannot reconstruct the order of ranks in the Parthian court from the 
titles found in inscriptions, usually in the frontier areas such as Dura Europos or Hatra, 
but several of the positions mentioned in the inscriptions may throw some light on 
the Parthian nobility. We find a general word hwtwy 'lord' possibly used as a general 
term of address for a member of the nobility. Those who were at court as friends and 
even bodyguards of the king probably had special designations such as nliwdr, the 
nohadares of Ammianus (14, 3), the 'first' or 'top friends' of the ruler. 65 These seem to 
have comprised a small top class of nobility, rather than the word being either a civil 
or military title. Civil and military titles, usually held by the nobility, will be 
mentioned below. 

In Armenia later the peasants seem to have been distinguished from the town plebes, 
and one may presume similar conditions in Parthian territories, but it is questionable 
whether the lower classes were subdivided into any kind of semi-legal ranks similar to 
the nobility. 66 In cities, especially those which continued their Hellenistic status of a 

62 J. Wolski, "Remarques critiques sur les Benveniste, [ch. 5, n. 74), 20, emphasizes the 
institutions des Arsacides," Eos, 46 (Wroclaw, correspondence between Parthian hwtwy and 
1954), 60, and K. H. Ziegler, Die Beziehungen Greek xvpios. The group of 'friends' of the 
zwischen Rom und dem Parlherreich (Wiesbaden, Hellenistic kings undoubtedly was paralleled in 
1964), 16. Parthian times. On the early predecessors of the 

63 M.-L. Chaumont, "L'ordre de preseances a la 'friends' see G. Widengren, Der Feudalismus, im 
cour des Arsacides d'Armenie," JA, 254 (1966). alien Iran (Koln/Opladen, 1969), 38-43. 
471-97. Undoubtedly one could interpolate from 66 Perikhanian, op. cit. [n. 56], 13, where the 
this list to a similar one in Iran under the Parthians, finakank' 'peasants' are distinguished from the 
called a gahnamak or 'book of ranks,' but details are ramikk' 'plebes', but Widengren, Der Feudalismus, 
lacking. 123, and his "Die Begriffe 'populorum ordo' und 

64 Cf. O. Seeck, ed., Notitia Dignitatum (Frank- 'ram'" Festschrift Walter Baetke (Weimar, 1966), 
furt/M, 1962, reprint of 1876 original), XXII for 384-87, claims that the ram were peasants on royal 
coats of arms, and 77-80 for the dux Mesopotamiae. domains and thus in a separate class, but his further 

65 Discussed with references by Harnack apud assertion (p. 387) that they possessed horses and 
Altheim, op. cit. [ch. 6, n. 49], 537-40. Note that followed the entourages of nobles is most question- 



The Parthians on the Plateau 221 

polis, differentiation according to professions did exist, but it does not seem to have 
been a social class division as among the nobility. When we turn to slaves and servants, 
the sources are silent, but some inferences can be made from words found in 
inscriptions, or in Armenian and later Middle Persian texts. The Armenian word 
ansahrik, which seems to be Parthian in form, originally meant a foreigner, probably 
taken in warfare, and it may have preserved this meaning throughout the Parthian 
period, for in the Sasanian period it probably fell together with the common word for 
any slave, bandag. 61 The legal position of slaves varied in different areas, for the 
Parthians did not institute a uniform system wherever they ruled but allowed local 
differences to exist. The word pariZtar (MP paristar), meaning a maid-servant, in the 
Parthian period referred to a hierodule, dedicated to service in a temple, but the 
extent of such 'slavery* is unknown either geographically or numerically. 68 The vt$ak 
or 'house slave' was universal but unfortunately nothing is known about them. The 
vineyard ostraca from Nisa and parchments from Avroman tell us nothing about the 
status of slaves or serfs in the dastkirt, the estates or patrimonies of the nobility. 69 
Slavery was widespread, especially in Babylonia, and while we hear of manumissions 
which made slaves free (azdt), the number of slaves was always large. 

The organization of the provinces was perhaps even more complicated than under 
the Seleucids and Achaemenids. The size of the satrapy had declined, but not so far as 
in the Sasanian period, when a satrapy was only a town with surrounding villages and 
lands. 70 The Parthians on the whole left the local lords and local administration 
intact when they conquered lands. In their homeland, however, the Parthian kings 
and nobility owned much land, as we see from the ostraca of Nisa, and royal estates 
must have been large here. Whether Parthia had a special position, free from taxes as 
Fars was at the beginning under the Achaemenids, is unknown but unlikely. 
Likewise, along the road from Khurasan through Rhages and Hamadan, i.e. Media, 
down to Seleucia, where Seleucid control had been centered, we may presume that 

able. His elaboration of the concept ram, as the meanings of dastkirt. For a summary of meanings 

corps of mounted peasants, is unconvincing in and references to previous articles, see Perikhan- 

"Recherches sur le feodalisme iranien," OS, 5 yan, op. cit. [n. 68], 458-60, where the religious 

(1956), 99. factor of dastkirt as a 'trust' or as a 'creation,' is 

67 Cf. A. G. Perikhanyan, "K voprosu o discussed. The Semitic mask for the Parthian word 
rabovladenii i zemlevladenii v Irane parfyanskogo is BN' (comp. Aramaic BNH) which implies the 
vremeni," VDl, no. 1 (1952), 13-27. On Armenia property of inheritance. See also G. Kh. Sarkisian, 
see S. T. Eremyan, "O rabstve i rabovladenii v "O dvukh znacheniyakh termina dastakert v 
drevnei Armenii," VDl, no. 1 (1950), 12-26. The rannykh Armyanskikh istochnikakh," in Ellenisti- 
use of the word bandak and bandaklfi 'slavery' in cheskii Blizhnii Vostok, Vizantiya i Iran, Festschrift 
Parthian could also mean something like 'submis- for N. Pigulevskaya (Moscow, 1962), 97-101, 
sion' to a king or a lord. Usage changed over the with much the same double meaning for the term 
centuries. in Armenia as in Iran. 

68 See A. Perikhanyan, Sasanidskii Sudebnik 70 It is clear that the hltrp or PHT ' of the Nisa 
(Erevan, 1973), 534-35, with further references. ostraca, like the dyzpty or 'commandant of a 
There were many kinds of dedications of 'slaves' to fortress' were not titles of great officials or nobles, 
temples, and this kind of slavery, of course, was not See I. M. Dyakonov and V. A. Livshits, Parthian 
ordinary slavery. Cf. P. Koschaker, Ober einige Economic Documents from Nisa, I (London, 1980), 
griechische Rechtsurkunden aus den dstlichen Rand passim. The fact that Classical sources continue to 
gebieten des Hellenismus, Abh. der Sachsischen AW, use the word 'satrap' for governors and also for 
42 (Leipzig, 1931), 76. almost any official, seems to indicate a generahza- 

69 Much has been written about the different tion of the term. 



222 Chapter VIII 

the Parthian king took over the role of the Seleucid ruler and appropriated the 
Seleucid crown lands for his own, so Media was probably ruled like Parthia. We are 
told that the Parthian Empire consists of eighteen kingdoms, according to Pliny (VI, 
112), eleven of which are 'upper' and seven of which are 'lower,' here meaning those 
on the plateau and seven in the plains of the 'Fertile Crescent.' On the plateau, other 
than Parthia and Media, there were probably several kingdoms subject to the 
Parthians in Armenia, Hyrcania, Azerbaijan (Media Atropatene), and possibly one in 
the mountains of Tabaristan. In the south we know of the kingdoms of Persis and 
Elymais, but Kerman may have been another independent area, with Seistan at some 
times under Parthian rule and in other periods independent or subject to rulers in 
India. Although eleven 'kingdoms' on the plateau cannot be identified, it is possible 
that Pliny is reporting accurate information. In the lowlands we know of the 
kingdom of Characene in the land called Mesene (Aramaic: Maisan) at the mouth of 
the Persian Gulf. North of Mesene was the central part of the land between the two 
rivers, ancient Babylonia, but called Beth Aramaye literally 'house of the Aramaeans' 
in Aramaic or Syriac and Asuristan in Parthian. This rich province, like Parthia and 
Media, was governed directly by the Parthian kings who maintained their winter 
residence on the plains, while in the summer they moved to the plateau. The great 
city of Seleucia had maintained a special status until Vologeses I built a competing city 
Vologesia to outrival Seleucia, which will be mentioned below. The area to the 
northeast of Seleucia, including the Diyala River basin and present Sulaimania, was 
called Beth Garmai in Syriac, Garmikan in Middle Persian, and the capital was 
Karkha de Beth Slok (in Syriac) modern Kirkuk. 7 ' This kingdom existed down to 
the coming of the Sasanians, when it was joined either to the central province of 
Asuristan or to the kingdom of Adiabene to the north. 

Adiabene, or Hadhyab in Syriac and in Parthian, called Nodshirakan by the 
Sasanians, from which Armenian Norshirakan is derived, was the land between the 
Greater and Lesser Zab Rivers with Arbela as its capital. 72 Little is known about 
Adiabene in Parthian times except the conversion of a queen of the country and her 
son Izates to Judaism, and later Izates was rewarded by Artabanus III for supporting 
his claim to the Arsacid throne by the grant of some land to the northwest, including 
Nisibis, to Adiabene. 73 In this extension of the domain of Adiabene, to the west and 

7 ' The Middle Persian form was reconstructed 'city made by the Natunians,' and later abbreviated 

by Marquart, op. cit. [ch. 3, n. 37], 21, and is to Shahrqird a site on the route from Baghdad to 

probably the glmykn, as well as the form glmykc'n Mosul. This seems correct although the location of 

in the Paikuh inscription ofNarseh, and on a seal of the town near Kirkuk, thus strictly not in 

the British Museum ; cf. W. B. Henning, "Notes on Adiabene, is curious The later deformation of the 

the Inscription of SapOr," in the Professor Jackson name in MP, using the name Ardashlr, founder of 

Memorial Volume (Bombay, 1954), 50. In addition the Sasanian dynasty, into Nod ArdashTr(akan) is a 

to the sources given by Marquart see J. F. Fiey, case of 'folk etymology' becoming official. Hen- 

Assyrie Chretienne, 3 (Beirut, 1968), 11-145. ning, "Mitteliramsch," [ch. 6, n. 57], 45. 
Kirkuk probably was founded by Seleucus, the 7i On this see J. Neusner, The Conversion of 

town of Seleucia of Pliny (VI, 117), while the Adiabene to Judaism," Journal of Biblical Literature, 

people, called Garamaioi, are noted by Ptolemy (I, 83 (1964), 60-63, and his "Shorter Note," in 

12,5 and VI, 1.2). Numeti, 13 (1966), 144-50. On the historical 

" In an inscription from Hatra, the name is geography of this area in the Parthian and Sasanian 

written ntwn , ky\ which J T. Mihk in "A propos period see L. Dillemann, Haute Mhopotamie 

d'un atelier monetaire d'Adiabene," RN, 3 (1961), orientate et pays adjacents (Paris, 1962), esp. 105-29. 
51-82, has identified as Natunia, plus iaro-kerta as 



The Parthians on the Plateau 223 

northwest, that land, called Beth Nuhadra in Syriac, with its center in the plain of 
ancient Nineveh, capital of Assyria, was absorbed. Neither under the Parthians nor 
under the Sasanians does this area of Beth Nuhadra, on the frontier of the Roman 
Empire, seem to have had a local dynasty; rather it was a military province governed 
by a Nohodar presumably a confidant of the king given a military command. In the 
flat land to the west of it was the desert and the province called Beth Ar(a)baye in 
Syriac (in Middle Persian Arbayastan and in Armenian, Arvastan) the chief town of 
which changed; at one time it was Nisibis. But at the end of the Parthian period, the 
kingdom and city of Hatra embraced the term Ar(a)baye. Hatra only expanded in the 
last century of Parthian rule and fell to the Sasanians in the last year of Ardashir's 
reign. 

Since the frontier fluctuated between the Romans and Parthians, the history of the 
rise and fall of minor principalities or the complicated changes in their boundaries 
cannot be followed. Other areas which in the Seleucid period were either provinces 
or separate principalities, and which may have preserved some independence into the 
Parthian period, were Sophene (Armenian Cop'k'), Zabdicene (Syriac: Beth Zabdai; 
Armenian: Zaudek'), Gorduene (Syriac: Beth Qardu; Armenian: Korduk') and 
others located in the mountain valleys to the north of Mesopotamia. 74 Since they 
were not integrated into the Parthian Empire but were usually under Armenian rule, 
or independent, they need not be further considered since they have little relation to 
the history of Iran except in the wars with Rome. Obviously there were many 
changes not only in the administration of various provinces and principalities. The 
revision of boundaries is difficult to follow, also on account of the changing fortunes 
of war in the land between the two rivers. 

As noted, the satraps were no longer the governors of huge lands such as Babylonia 
in the Achaemenid period, and we may assume that the subdivisions of the former 
large territories were not governed by officials called satraps, but the governors of 
large provinces, in continuation of the Seleucid tradition, were called strategos in 
Greek. In Parthian the equivalent term was probably hhrdry or shahrdar, with a 
general meaning of 'holder of the realm' or 'sovereign.' Under the governors were 
satraps, as we see from the Nisa ostraca. There were many officials under the satrap, 
especially accountants to care for the revenues, '/imrfer, the hamarkar. The chief 
collector of taxes was an important official called hrkpty, or \kpty and hrgwpt in 
Parthian, an office formerly mistakenly interpreted as argbad or 'fortress 
commander.' 75 For the Parthian period we have no information about the position of 
the chief tax collector in the hierarchy, but presumably it was not high and only 
under the Sasanians does the office gain in importance. 

74 Names of Seleucid provinces, however, do " Much has been written about the argbed; cf. 

not tell us whether local princes ruled in them or my remarks apud C. B. Welles, The Parchments and 

Seleucid governors. The area about Nisibis was Papyri, The Excavations at Dura-Europos (New 

called Mygdonia, after the river, by the Greeks Haven, 1969), 111-12, to be corrected and 

but does not seem to have been an independent superseded by the above; also Szemerenyi, op. ctt. 

principality, and the same is true of Apolloniatis on [ch. 4, n. 63], 366-75. 
the Diyala River, nearby Sittacene, and others. 
Information on native names and changes in 
administration is extremely sparse. 



224 Chapter VIII 

When we turn to the army, our sources are also deficient, and the temptation to 
turn to earlier Achaemenid or later Sasanian times to reconstruct the Parthian army 
should be resisted, for the Achaemenid organization broke down with the ever 
greater use of mercenaries, also a feature of the Seleucid kingdom. We may assume, 
however, that some sort of decimal organization of the army continued under the 
Parthians, and that there was an army commander, spdpty or spahbad, especially in 
time of war, and that the cavalry, the 56 Vy, was especially important, as we know 
from Classical sources. In the frontier areas, such as Nisa, we find the military titles of 
mrzwpn 'margrave' and dyzpty 'fortress commander,' the former probably the officer 
in charge of the frontier troops, while the latter was the officer in charge of a fort, as 
the name says. 76 From Classical sources we learn that the cavalry tactics of attacking 
and then feigning retreat, with the famous arrow shot, the 'Parthian shot,' turning in 
the saddle while fleeing, made a great impression on the Romans. When Justin (XLI, 
2) says that fifty thousand cavalry opposed Marc Antony, of which only four 
hundred of them were free men (liberi), he means that the nobles only numbered so 
many. Although it may have seemed to outsiders that the common soldiers were like 
slaves, they were more like followers or even serfs of the nobility. The feudal nature 
of Parthian society must have evolved throughout the course of history, but there is 
no evidence of the peculiarly Western European form of vassalage, and all that went 
with it. The Parthian form of 'feudalism' seems simpler, with the followers, or the 
peasants in villages belonging to a lord, supplying troops when needed. Sources for 
further speculation are lacking. 

A loan document from Dura-Europos may throw some light on Parthian political 
and military provincial organization. It gives the titles and honorifics of two officials, 
one military and one civil, but it must be remembered that Dura was located on the 
frontier, even though the document itself is not directly from the frontier area. A 
certain Phraates, a eunuch, and hargbad or 'tax collector,' and 'one of the people' (at the 
court of, or in the entourage of) Manesus, son of another Phraates, who was governor 
of Mesopotamia, Parapotamia as well as Arabarkhos, or 'ruler of the Arabs,' made a 
loan to another person. One of the witnesses was Metolbaessa, commander of the local 
garrison. The governor held not only the civil office, but had another title which is 
damaged in the parchment, which was restored as one who receives taxes, but this is 
uncertain. 77 In any case, both the commander of the garrison and the governor carry 
honorifics, the former belonging to the (order of the) 'first and chiefly-honored 
friends and bodyguards' (presumably of the king), while the governor belonged to 

76 For the titles and bibliography, see the 11 M. I. Rostovtzeff and C. B. Welles, "A 

glossary of P Gignoux, Glossaire des Inscriptions Parchment Contract of Loan from Dura-Europus 

Pehlevies et Parthes (London, 1972). The title on the Euphrates," Yale Classical Studies, 2 (New 

ptyktvspn'paygospan' or 'padhospan' is uncertain, for Haven, 1931), 51. The restored title TTapaXr)Trrr]s 

in Sasanian times it seems to have meant a 'gatherer for one's self,' has not been found 

'governor-general' of a number of provinces, or elsewhere. If the word is not Greek, it might be a 

the officer in charge of the armies of one of the four parallel to slrategos, like PHT', but I have no 

quarters or frontiers of the empire In Parthian suggestion which would fill the gap in the 

times it may have meant the same as 'governor,' or parchment. Cf. H. Bengtson, Die Strategic in der 

more likely the same as marzban, 'warden of the Hellenist. Zeit II 2 (1964), 279-280: irapa.7Ta.TTOv 

marches,' but we cannot determine thejurisdiction (Enns-lin and Mlaker). 
of the official. 



The Parthians on the Plateau 225 

the (order of the) batesa and the azatan or nobility. 78 Just what batesa means is 
uncertain, but I would like to suggest that it does not mean the title byfhS (bidakhsh) 
with many forms, but is rather related to the Middle Persian word p'thfdy 
(padehshay) 'having rights by authority,' a civil honorary order of nobility. 79 The 
bidakhsh is more military in character than civil. Much has been written about this 
title found in many sources, and no matter what the etymology, the historical 
importance of this title in Parthian times was not that of a 'second king,' or second in 
line of succession, but rather the representative of the king, and since we find the title 
mostly used on the western and northern frontiers of the Parthian domains, we may 
conjecture that this official was originally the king's representative at the courts of the 
sub-kings or 'vassal' rulers. From later usage in Armenia and Georgia the term may 
have already developed in Parthian times to a meaning of 'warden of the marches,' or 
something similar to the mrzwpn (marzban), although the functions of both offices 
were undoubtedly more than simply military commands. 80 It is virtually impossible 
to distinguish official designations from popular usage or synonyms in nomenclature, 
as we have seen with the Parthian words for 'prince,' 'lord,' and others. A word such as 
prmtr (framatar) 'commander,' which was civil, however, rather than military, gives 
no clue to his functions, but we may conjecture it was the equivalent of modern 
Persian-Arabic ra'fs, or 'director' of any civilian institution. As noted, the values or 
meanings of titles changed over centuries, and what may have existed in early 
Parthian times frequently was either a memory or altered in significance by the time 
of the Sasanians. Certain constants can be recognized, however, and one may postulate 
the continuous existence of a village chief under the Parthians as the basic office of 
authority. 81 Above the village chief was a district or provincial chief of the rwdyst'g 
(NP rOsta) who was called a 'satrap' (hs'trp or P^iT'), while above the satrap was the 
ruler of the hhr, the shahrdar, the local potentate, corresponding to the Achaemenid 
satrap and the Sasanian dstandar or head of a large province. 82 When we come to the 
Sasanian period more sources aid us, but for the Parthians these three divisions in the 
hierarchy of civil jurisdiction were the most widespread, although, as usual, 
exceptions probably did occur under the Parthians as at other times. 

78 Cf. Welles, op. cit. [n. 75], 115-16. My side' (see n. 76) may have been a synonym for 
remarks in "Some early Iranian titles," Oriens, 1 5 marzban, but in Parthian times we have no sources, 
(1962), 352-54, are still valid regarding parallels and a reading backwards from the Sasanian period 
and honorifics as well as offices in spite of attacks may be misleading. 

by F. Altheim/R. Stiehl, Die Araber in der alten 81 The Parthian word for 'village' is found only 

Welt, 1 (1964-69), 635-38, or Szemerenyi, op. cil. m the Aramaic logogram QRYT' in the Nisa 

[ch. 4, n. 63], 371. They mistakenly assumed ostraca, which has been interpreted as *dyz(+pty) 

parallels to mean identities or synonyms, whereas lit. 'lord of the citadel' (dizbed), but it may have 

one, Metolbaessa, holds a military office, while the been rather *wyspty or *visbed 'village chief,' even 

other, Manesus, is a civil official. though every village may have had a wall around 

79 The Parthian form would be *pdyj?s"h- which it like a citadel. 

the Arabs settled around Dura might have 82 In some areas the chief of a district may have 

pronounced *badesa. The Greek form implies a held the title pthV (padehsha), as Mar Qardaq who 

foreign word, and the uninfected plural would was the padehsha of the district of Athur (Assyria) 

indicate an (honorary) title rather than an office, in Sasanian times; cf. P. Bedjan, Acta Martyrum et 

although it might mean 'district officer,' on which Sanctorum, 2 (Leipzig, 1890), 445, line 2. Or this 

see infra n. 82. word could be only a general designation of one in 

80 Likewise the ptykwspn or 'guardian of the authority. 



226 Chapter VIII 

The local potentates were on the whole Parthian princes, especially after 
Mithradates II when members of the Arsacid family were installed as local rulers in 
many areas. 83 The great families and rulers of the Parthians were traditionally seven, 
the two most prominent of which were the Suren, who maintained estates in Seistan 
and elsewhere in Iran, and the Karen family primarily in western Media with a center 
in Nihavend. 84 The other great families are not mentioned in any source of the 
Parthian period but notices from Sasanian times may be referred to the early dynasty. 
One family probably held the office of army commander of the Parthian forces for 
several generations, such that it became known by the title 'sp'dpty (Aspabad), and 
their chief seat apparently was in Khurasan and Gurgan. The later Sasanian families 
may have existed in Parthian times, or they may have new names; we do not know. 
One family, the Mihran, reputedly had its center at Rayy, but another name 
Spandiyad is also connected with Rayy, while still another Varaz seems to be a new 
noble family coming to the fore in Sasanian times. We may postulate the existence of 
several, perhaps seven according to tradition, high noble families of the Parthians, but 
we have no information about them. 85 Undoubtedly there were protocols and a 
hierarchy of rank among families at the court of the Parthian Arsacids, as we find in 
Armenia, but again the details elude us, although, as noted, Tacitus (Annals, VI, 42) 
says that the right of crowning the Arsacid king was a prerogative of a member of the 
Suren family, and there were surely other similar protocols. 86 The king as the chief of 
the nobles had the right of assignment of offices or fiefs, provided that traditions were 
maintained. To return to the question of a senate of Parthian nobles who had the right 
to banish or elect kings (Justin XLII, 4) (or as Strabo [XI, 515] says there were two 
assemblies, one composed of relatives of the king [the nobility] and the other of Magi 
and wise men), undoubtedly the great families had much influence in such meetings, 
but if one or two assemblies existed, they were related to the tribal traditions of the 
Parthians rather than to well-established governmental institutions. 87 The feudal 
relations between sovereign and nobles can only be inferred from later practices, but 
we may suppose the existence of small courts copying the royal court, coats of arms, 
and such accoutrements of what we know from Western 'feudalism,' under the 
Parthians. 

Cities seem to have flourished more in the Parthian era of history as compared with 
the Seleucid or Sasanian periods, probably because of the expanding east- west trade 
rather than any liberal policy of the Parthian government. Seleucia has been 
excavated and the continuing Hellenistic institutions and traditions indicate that 

83 See esp. Tacitus, Annals (XII, 14 and XV, 2), clan.' In Manichaean MP we find zndbyd 'head of a 
where various Arsacid princes are mentioned as zanm or tribe,' which may have replaced an earlier 
installed in local kingdoms. title of *nafapat. On the Latin term megislanes see 

84 See J. Marquart, "Beitrage zur Geschichte the discussion in T. Mommsen R&mische Ge- 
von Iran," ZDMC, 49 (1893), 635-36, for schichle, 5 (Berlin, 1885), 343, n. 2. 
references, also Herzfeld, "Sakastan" [ch. 7, n. 28], 86 On Armenia see Chaumont, op. cil. [n. 63], 
64-66, where his derivation of the Karen family 471-97. This ga\f namak, or notitia dignitatum, is 
from a governor installed by Mithradates I is from a later period but is derived from Parthian 
highly conjectural. practices. 

85 It is unclear whether the chief of a great 87 Cf. G. Widengren, op. cil., [n. 65], 108-115, 
family would have been called a *nafapat in with further references. 

Parthian whence Armenian nahapet, 'chief of a 



The Parthians on the Plateau 221 

Parthian rule brought little change from Seleucid times. The example of the great 
metropolis in the lowlands may have been unique, however, since the sources on 
other towns are few and tell us little. The Parthians seem to have neglected the 
lowlands, or at least southern Babylonia, during the pre-Christian period, since 
archaeological surveys show a neglect of agricultural land there. 88 Compared to the 
Sasanian period, when land under cultivation and irrigation greatly increased while 
urban life did not, for the Parthians, at least in Mesopotamia, the reverse seems to have 
obtained. After the death of Mithradates II internecine struggles between Arsacid 
contenders for the throne were not conducive to the flourishing of cities. Pliny (VI, 
122) says that in his time the city of Seleucia on the Tigris was free (libera) and 
retained its 'Macedonian' manners. Seleucia, as undoubtedly other cities too, revolted 
against Parthian control from time to time, but the sharply delineated preferences of 
the Hellenic aristocracy' as opposed to the native 'masses' have been perhaps too much 
emphasized by modern scholars. 89 Seleucia supported Gotarzes against his brother 
Vardanes, resisting the siege of the latter for seven years after a.d. 36, according to 
Tacitus (XI, 9). Although one can distinguish between three groups — the Parthians, 
Greeks and natives — to consider the inner conflicts of the three as the key to urban 
revolts may be too simplistic a surmise. Numismatic evidence suggests that the city 
after a.d. 24 ceased to issue its own local bronze coinage. The old Seleucid 
organization was maintained there and, on the whole, in other 'polis' cities, if one is to 
judge by the Greek letter of Artabanus III to the city of Susa; for a city council and 
various magistrates governed the city undoubtedly continuing the old organizations, 
although the influence of Parthian royal officials on the city must have been great. 90 
Ctesiphon, a suburb of Seleucia, across the Tigris River on the east bank, was the 
residence of the Parthian kings, according to Strabo (XVI, 743), although Pliny (VI, 
1 22) claims the city was founded to draw the population away from Seleucia, an echo 
of the founding of Seleucia itself vis-a-vis Babylon. Another city, Vologesocerta, was 
founded by King Vologeses nearby, according to Pliny, although which king this was 
is uncertain. Likewise the location of this city and its identification with the 
Vologesias of Ptolemy, Ammianus, and others, is in dispute. 91 The Parthian kings, 

88 R. M. Adams and H. J. Nissen, The Uruk 90 F. Cumont, "Une lettre du roi Artaban III," 
Countryside: the Natural Setting of Urban Societies CRAI (1932), 238-60, and Welles, Royal Corre- 
(Chicago, 1972), 57, and H. Nissen "SUdbabylo- spondence, [ch. 6, n. 68], no. 75. For a discussion of 
nien in parthischer und sassanidischer Zeit," the roles of the archons, the epistatts and satrap, see 
Baghdader Mitteilungen, 6 (1973), 82. G. A. Koshelenko, "Gorodskoi stroi polisov 

89 Cf. C. C. Hopkins, ed., Topography and zapadnoi Parfii," VDI, 4 (1960), 79-80. Ecbatana 
Architecture of Seleucia on the Tigris (Ann Arbor, continued as a summer capital of the Parthians, 
1972), 160, where the dichotomy of natives according to Strabo (XI, 522). 

supporting rebels and Greek commercial oligarchy "A. Maricq, "Vologesias, l'emporium de Ctesi- 

favoring the Parthians, or the Romans if powerful, phon," Syria, 36 (1959), 271, maintained the 

is maintained. Cf. McDowell, Coins op cit., [n. 27], identity of the two, situated on the 'Royal Canal' 

226-27. It is possible that after the revolt masses of between the Tigris and Euphrates, while N.-L. 

non-Greeks settled in the city, as suggested by G. A. Chaumont, "Etudes dTiistoire Parthe III. Les villes 

Koshelenko, "Arkhitektura zhilishcha Greches- fondees par les Vologese," Syria, 51 (1974), 81, 

kikh gorodov Parfii," in Antichnyi Gorod (Mos- places Vologesia near later Kufa and Vologeso- 

cow, 1963), 181. Tacitus (Annals, VI, 42) men- certa on the canal, 
tions the senate of Seleucia and the conflict 
between the aristocracy and the common folk of 
the city. 



228 Chapter VIII 

however, were not founders of numerous cities as were the Sasanians, and the cities 
which flourished on the trade routes such as the capital of Characene, Hatra and 
Palmyra did so because of Roman demand for eastern spices and luxuries rather than 
from Parthian support of them. They also flourished because of their relative 
independence from the two great powers. 

Whereas the history of the trading cities or 'caravan' cities has been treated as a 
continuation of Hellenism by modern authors, the political ideology of the Parthians, 
on the other hand, developed independently of the cities. Since the Parni came from 
the steppes of Central Asia, we may assume that they had only oral traditions and no 
written history. Likewise the inhabitants of the province of Parthia which they 
conquered, unlike the Persians of Fars province, probably had few significant 
traditions or memories of the Achaemenids. There is no evidence that the early 
Parthians paid much attention to any political ideology, other than those of the 
ancient Iranian or even Indo-Aryan or Indo-European tribes. With their conquest of 
settled Iranians and of descendants of the Greeks and Macedonians of the Seleucid 
Empire, two traditions were added to their own. The Greek or Seleucid tradition was 
accepted by the Parthian rulers, as we can see from the title 'philhellene on the coins of 
most of the rulers from the beginning to the end of the Arsacids. From the coins, at 
least, there is no evidence of an anti-Hellenic sentiment, and there was little reason for 
this among the Parthians. The old Iranian or Achaemenid tradition is more difficult 
to grasp, since evidence for the promotion of dynastic links with the past, great 
dynasty is scanty. The Parthians, of course, fought against the Seleucids, but the 
treatment of Demetrius II by Mithradates I indicates that the Parthians were not 
simply partisans of Iran versus the enemy, Greece. As at Commagene, the Parthians 
may well have created a mixed Persian-Greek fictitious genealogy, as Wolski 
suggests, but this is only surmise. 92 Arrian (Parthika, frg. 1 in Syncellus) says that the 
Arsacid family was descended from the Achaemenid Artaxerxes, probably the second 
of the line, whose name before becoming ruler was Arsaces. From the title 'king of 
kings' on coins of Mithradates II and from indications in Tacitus (Annals, VI, 31) and 
others, the belief that they were the heirs of past Persian glory and empire was 
probably promulgated by the Arsacids after the first century B.C. 93 

Perhaps too much has been made of the reappearance of the title 'king of kings' and 
the appearance of Aramaic letters on the coins of Vologeses I to support the hypothesis 
that the Parthians adopted an anti-Greek attitude by the first century of our era, with a 
corresponding exaltation of Achaemenid traditions. 94 The use of Aramaic at Nisa 
instead of Greek is no real evidence for anti-Greek feeling, for from Avroman in 
Kurdistan documents in both Greek and Aramaic were found and obviously in some 
areas Greek was more used than Aramaic, or vice versa in other places. The Aramaic 

92 J. Wolski, "L'ideologie monarchique chez les Nachkben, ed. by R. Stiehl, 1 (Berlin, 1969), 321. 
Parthes," in Studi vari di Storia greca, ellenistica e See also F. Altheim, Literatur und Gesellschaft im 
romana (Milan, 1977), 233, and his "Les Achemen- ausgehenden Altertum, 1 (Halle, 1948), 83. 

ides et les Arsacides," Syria, 43 (1966), 74-77. 94 Especially by Wolski in "Les Parthes et leur 

93 Cf. J. Neusner, "Parthian Political Ideology," attitude envers le monde Greco-Romain," Dia- 
IA, 3 (1963), 56-58, and Wolski, "Les Achemen- logues d'histoire ancienne, 2 (Besancon, 1976), 284- 
ides" [n. 92], 87, and his "Arsakiden und Sasani- 85. Contrary to Wolski, Greek legends do not 
den," in Beilrage zur alien Geschichte und deren disappear from Parthian coins although they 



The Parthians on the Plateau 229 

legends on Parthian coins have been misunderstood. They are not legends, for those 
are in Greek, albeit debased, to the end of the dynasty. The Aramaic letters are rather 
mint, or mint masters' marks, and names appear on coins in Aramaic only in the last 
century of Parthian rule, which should be attributed to the decline of knowledge of 
Greek rather than a conscious anti-Greek policy on the part of the Arsacid rulers. The 
question of the use of Aramaic, however, has many ramifications. 

Under the Seleucids Greek was the official language and script of the state. Parallel 
to it went Aramaic, which continued to be used as the principal means of written 
communication in areas away from the main road from Mesopotamia to Bactria. 
Thus in Parthia, Persis, Azerbaijan, Central Asia and the Caspian districts, Aramaic 
continued to be used with little or no inroads by Greek. As time continued the usages 
of Persis varied more and more from those of Sogdiana and elsewhere. Local scribes 
began not only to use more Iranian words in their texts but also to introduce local 
grammatical features. All the time the texts were read aloud in the local language by 
the scribe, whose duty was to write messages for government officials, the local ruler 
or for anyone who paid him to write for them. The ostraca from Nisa, dating into the 
first century a.d., can be read as Aramaic with Iranian words and even endings, or 
they may be read as Parthian, written in ideograms or logograms, but read aloud in 
Parthian alone. For the inhabitants of Nisa were Parthians who spoke Parthian and 
obviously did not speak Aramaic, not even a broken form of it. In Persis the Aramaic 
legends on coins were read aloud as Persian, in Georgia and Armenia, similar 
inscriptions were read aloud in the local language. The fully developed Parthian 
system of writing, however, which may be characterized as an Iranian text with 
Aramaic ideograms in it, as contrasted with the earlier Parthian writing in Aramaic 
with Iranian words and some endings, is only attested from the third century a.d., the 
end of Parthian rule. The second century is blank, for we have no Iranian inscriptions 
from this period. We do have Semitic inscriptions from this period, however, in 
Hatra, Palmyra, and even from Characene and Elymais, but none from Iran. One may 
conjecture either a gradual change from the first to the third century A.D., or possibly 
a conscious effort to reform the writing at some date during this period. The separate 
development of local scripts in Mesopotamia, in Georgia, Khwarazm and Sogdiana is 
part of the overall decline of the use of Greek and the relative isolation of the 
peripheral areas of the Iranian plateau. More evidence for these local developments 
can be found in the Sasanian period when both religions and scripts proliferate, giving 
more material for a reconstruction of the history of both scripts and languages. In any 
case, there is no evidence for either an anti-Greek language/script or an anti-Aramaic 
language/script movement; rather time took its course in both cases. 

The view of religion under the Parthians is filled with contradictions, for again we 

become debased, while his further remark that the norm. The change in the chancellery of Nisa in the 

first Arsacids borrowed the model, even the idea of middle of the first century B.C. is revealed in the 

coinage from Persis is unconvincing. The Aramaic smaller, more legible style of writing on the 

letters KN, which appear on the reverses of several ostraca, as well as the mention of scribes, according 

rare Susa 'type' coins of Seleucus 1, and the coins to I. M. Dyakonov and V. A. Livshits, Dokumenty 

with legend WHSW/R are enigmatic but they do iz Nisy (Moscow, 1960), 17, but their significance 

not change the general picture of Seleucid and escapes us. 
Parthian coinage in which Greek legends are the 



230 Chapter VIII 

have no written sources from this period of Iranian history and archaeological 
evidence is equivocal. From this evidence, theoretically, one may distinguish between 
Greek cults and influences, local beliefs and cults, and Zoroastrianism on the Iranian 
plateau under the Parthians. In the east, Buddhism and various Hindu cults are 
attested at various times and in various places, while in Mesopotamia Judaism and 
various local religions, some of them descendants of ancient Babylonian religions, 
existed. The farther east and Mesopotamia will be discussed in later chapters, while 
here the emphasis will be on the religious situation in the heart of the Parthian 
domains. 

Continuing Greek influences, and possibly cults which were Greek in origin, are 
attested by inscriptions and art remains. Herakles undoubtedly was a popular hero- 
deity in Iran in Seleucid and Parthian times with some sort of a cult and followers. 
The Greek inscriptions in a cave at Karaftu, about 20 km. west of Takab on the 
Azerbaijan— Kurdistan border, with the apotropaic message, "here lives Herakles; 
may no evil enter," from the Seleucid period, plus another on the rock relief of 
Herakles Kallinikos at Behistun from 149-148 B.C., plus several statues of the hero- 
god, all attest his popularity in Iran. 95 Throughout the Seleucid and Parthian periods 
a process of amalgamation or syncretism continued, whereby local deities were 
identified with Greek deities. Thus in Iran, Herakles probably was early identified 
with Verethragna, later Bahram, Vahagn in Armenia, and also in Commagene on the 
royal statues and inscriptions of Antiochus, the local ruler. 96 On coins and in the arts, 
evidence of the assimilation of Greek and Iranian deities indicates a widespread 
acceptance of this syncretism, which is not unexpected since both peoples came from 
the same Indo-European linguistic family. Likewise the fire cult of Iran was matched 
by a Greek counterpart, and the word ESTIAZ on a rhyton from Nisa could refer 
to a combined fire cult. 97 In Iran, of course, the fire cult was characteristic of the 
Zoroastrian religion then as later, and when Isidore of Charax (par. 1 1) speaks of an 
ever-burning fire in the city of Asaak where the first Arsacid king was crowned, we 
have the prototype of later Sasanian practice, where in the royal fire temple a fire was 
started at the beginning of the reign of a new king. Ancestor worship, or rites in 
honor of the frauasis or spirits of one's ancestors, was a feature of the religion of the 
Parthians, as later in Iran. 

Popular beliefs differed from more formal religions, and worship of the elements, 
as well as the stars and the sun, apparently was widespread. 98 Likewise the 

95 H von Gall, "Die Kultra'ume in den Felsen of the Parthians (Bombay, 1925), 21-22. On some 
von Karaftu bei Takab," AMI, 1 1 (1978), 94; and coins of Mithradates 1 a standing Herakles appears 
for Behistun, L. Robert in Gnomon (1963), 76. on the reverse. 

Many terra-cotta figurines from Seleucia indicate 97 G. A. Koshelenko, "Grecheskaya nadpis' na 

the popularity of Herakles there; cf. Frye, supra, Parfyanskom ritone," VDI, 2 (1967), 167-70. His 

Heritage, 170-76, with references. identification of the Greek deity Hestia with Vesta 

96 F. K. Dorner u. T. Goell, Arsameia am in Rome and Agni in India is questionable since the 
Nymphos, Istanbuler Forschungen, 23 (Berlin, Greek Hestia although patroness of the hearth does 
1 963), 223 ; and for Vahagn see G. D. Bardumyan, not seem to have had a special fire cult dedicated to 
"Gosudarstvennye religioznye kulty Armyan," her. 

Vestnik Moskovskogo Vniversiteta, 2 (1 976), 87-88. 98 See C. Colpe, "Die Bezeichnung 'iranisch' fur 

His etymologies must be viewed with skepticism. die Religion der Partherzeit," ZDMG, Supple- 
See also J. M. Unvala, Observations on the Religion mentband (1969), 1013-14. The popularity of 



The Parthians on the Plateau 231 

identification of certain deities with planets or stars existed, but the significance of 
astrology in popular belief is uncertain, for just what an 'identification' of the sun with 
Mithra or Venus with Anahita meant is unclear. Anahita in Sogdiana was 
distinguished from another female deity Nane or Nanaia as also in Armenia, but the 
relation of these two deities to each other and to the old belief in a mother goddess are 
both unknown." In any case, syncretism and the identification of deities with each 
other, with a common cult or set of rituals, was widespread. Whether one cult was 
more popular or more followed in one place than another, or whether one cult was 
more patronized by the ruler or aristocracy than another, cannot be determined but is 
likely. If we try to assemble the notices about the religion of the Parthians, many 
popular practices are found, such as the remark of Josephus (Jewish Antiquities, XVIII, 
344) that the Parthians carried small idols with them on voyages, or that trees were 
worshipped in Mesopotamia in Parthian times. 100 Popular beliefs certainly may be 
found, but what of an 'official' religion - of Zoroastrianism? Apparently no official 
state religion existed, as later under the Sasanians, but the Parthians have suffered in 
later sources as being poor Zoroastrians or even non-Zoroastrians. 

Undoubtedly there were Magi who condemned the presence of statues in cult 
places while others did not object; or some adhered to one set of rituals and others to 
another. 101 Likewise some indulged in time speculation (Zervanism) while others 
did not, but again there is no evidence for formal sects or religions under the 
Parthians. The extension of Iranian cults, especially one of fire with Magi, into 
Anatolia is mentioned by Strabo (XV, 733) and others, but one may conjecture that 
this was the legacy of Achaemenid domination. At the end of the Parthian period, 
however, Zoroastrianism was developing into an organized religion in competition 
with other religions such as Judaism, Christianity and possibly the one we know as 
Mandaism. According to a late Middle Persian text, the Denkart, a Parthian king 
Valakhsh (Vologeses) ordered a gathering of all oral texts of the Avesta, and any 
teachings derived from it, presumably in order to codify the sacred texts of the 
Zoroastrians, but which king of this name is intended cannot be determined. 102 At 
least, this is an indication that the Parthians were not indifferent to the Zoroastrian 



shamans in this period (p. 1017), however, is ' 00 See E. Sachau, trans.. Die Chronik von Arbeta, 

nowhere attested, even though not impossible. Abh. PAW (Berlin, 1915), 52. The reliability of 

Buddhism in the eastern part of Parthian domains this Syriac source has been questioned, but it is 

will be discussed in the next chapter. hardly a modern fabrication. 

''"On Anahita see H. Lommel, "Anahita- ,01 See J. Bidez and F. Cumont, Les mages 

Sarasvati," in Asiatica, Festschrift Fnednch Weller Hellenish, 1 (Paris, 1938), 74. Every female figure 

(Leipzig, 1954), 405-12, with references. On on rock reliefs, silver bowls, or on seals is usually 

Sogdiana see N. V. Dyakonova and O. I. Smirnova, identified as Anahita, but this is too simplistic since 

"K voprosu o kulte Nany (Anakhity) v Sogde," not every representation is a goddess, and Anahita 

VDI, 1 (1 967), 74-81 . Isidore of Charax (6) speaks was not the sole female deity as even a reading of L. 

of a temple of Artemis at Concobar (Kangavar), Gray, The Foundations of the Iranian Religions 

which has been identified with the massive ruins in (Bombay, 1 929), indicates, 

that town. Anahita also has been identified with l02 Cf. the text of D. M. Madan, 1 (Bombay, 

Cybele in Asia Minor, and probably elsewhere 191 1), 412, 5-10and trans, by Zaehner.op. cit. [ch. 

with other deities. Cf. also M.-L. Chaumont, "Le 3, n. 40], 8. 
culte de la deese Anahita (Anahit) dans la religion 
des monarques d'lran et d'Armeme au HI e siecle de 
notre ere," J A, 253 (1965), 167-71 



232 Chapter VIII 

religion, and continued to support its development. The names from the Nisa ostraca, 
such as Sroshdat, Tirdat, Vahuman, and many others are clearly Zoroastrian, but 
whether the large number of names with Mithra in them, such as Mihrfarn and 
Mihrdat, point to the existence of a separate cult devoted to Mithra, the prototype of 
Roman Mithraism, cannot be determined. 103 Whether such Zoroastrian texts as the 
Vendidad were composed in Parthian times cannot be determined although not 
unlikely. In any case, Zoroastrian practices, such as next-of-kin marriages, exposure of 
the dead, are attested for Parthian times, indicating a continuity from the past and a 
link to the Sasanians. 

At the same time, however, the later tradition that the Arsacids were lax in their 
religious zeal seems correct, for in this manner the Parthians continued Achaemenid 
tradition. With the rise of Christianity and other organized religions, however, the 
Parthians had to combat conversions from Zoroastrianism which probably caused a 
tightening of the organization and priesthood of Zoroastrianism. Jews, on the whole, 
were on good terms with the Parthians, especially after the Roman occupation of 
Palestine. 104 The episode of the two Jewish brothers Anileus and Asineus, as reported 
by Josephus {Jewish Antiquities, XVIII, 9), who created a private army and domain in 
Babylonia in the time of Artabanus III, indicates the importance of the Jewish 
settlement there as well as the weakness of the central government. 

We have mentioned the disposal of the dead by exposure, but this practice had not 
become general among the Parthians as in the Sasanian period. In Merv and elsewhere 
in Central Asia burials in large jars have been found by archaeologists but astodans or 
ossuaries for the bones have also been found. 105 The principle of non-pollution of the 
soil is maintained in both cases, and different forms and styles of coffins have been 
found in graves of the Parthian period in Mesopotamia as well as on the plateau. Thus 
evidence for the Zoroastrian faith of the defunct should not be limited solely to bodies 
exposed to vultures. Obviously in the pre-Sasanian period there was no prescribed 
manner for the disposal of the dead, and only the general admonition not to pollute 
the elements prevailed. The ancient Indo-European practice of cremation which we 
find among the Greeks and other Indo-European peoples evidently had vanished 
from the Iranian plateau by this time, since there is no evidence for ashes of the dead in 
Parthian burials, although it is conceivable that some people followed the Greek 
practice for one reason or another. In any case, the Zoroastrian abhorrence of 
polluting the fire already held sway in Parthian times. We now know that much of 

103 I. M. Dyakonov and V. A. Livshits, Teslamenlum, Supplement 4 (Leiden, 1957), 197- 

Dokumenty i> Nisy (Moscow, 1960), 24. The 240. 

curious appearance of the name Sasan apparently as l05 G. Koshelenko and O. Orazov, "O pogrebal- 

a deity, in the Parthian ostraca from Nisa, in nom kulte v Margiane v Parfyanskoe vremya," 

compounds such as Sisanbokht, Sasandat, indicates VDl, 4 (1965), 42-56. In the footnotes further 

a wider pantheon, or even a cult of divinized references to ossuaries and other burials may be 

ancestors, than we know from later times. See found. See also Y. A. Rapoport, "Some Aspects of 

V. A. Livshits, "Parfyanskii reonim Siwart," P/$me/!- the Evolution of Zoroastrian Funeral Rites," and 

nye pamyatniki i problemy istorii kultury narodov G. P. Snesarov, "The Mazdeist Tradition in the 

vostoka, ed. by P. A. Gryaznevich (Moscow, Burial Customs of the Peoples of Central Asia," 

1977), 93-97. in Trudy XXV Mezhdunarodnogo Kongressa Vos- 

1 0i See G. Widengren, "Quelques rapports entre tokovedov, 3 (Moscow, 1963), 127-40. 
juifs et iraniens a lepoque des Parthes," Vetus 



The Parthians on the Plateau 233 

the culture and practices of the Sasanians was Parthian in origin and the former have 
received credit for innovation, when they really were building on the past of their 
enemies whom they successfully sought to denigrate. 

WARS WITH ROME 

The Classical sources on Parthia naturally deal mostly with the wars between the 
Romans and Parthians, although some information about internal affairs or dynastic 
quarrels is recorded. The accounts are, of course, partisan and must be used with care, 
but at least the general course of the struggles between the two powers and the 
outcomes of such contests are usually reliable. On the whole, the Romans had the 
upper hand, especially in the time of the empire when Parthian central authority had 
been weakened and local rulers had asserted more authority than in the first two 
centuries of Parthian rule. Phraates III saw the fall of Tigranes II the 'Great' of 
Armenia and Mithradates of Pontus to superior Roman arms and the generalship of 
Lucullus, but it was not until Pompey replaced Lucullus that the Romans were able to 
impose a peace on the area. The Classical sources on the diplomacy and intrigues 
between Tigranes the younger, son of Tigranes the Great of Armenia, and Phraates 
have been well summarized by Debevoise, making repetition here unnecessary. 106 
Tigranes and Phraates composed their differences after their ambassadors met 
Pompey in Syria in 64 B.C., and it seems the boundary between the Armenian and 
Parthian states was drawn between the northern border of Adiabene and Nisibis, 
while Syria was annexed by Rome. Phraates was murdered in 57 B.C. by his sons 
Mithradates and Orodes (wrwd), and the former apparently seized power only to be 
ousted by his brother, who after much fighting succeeded in capturing and executing 
Mithradates in 54 B.C. The following year the new Roman governor of Syria, Crassus, 
with overweening arrogance, attacked the Parthian domains and lost his life as well as 
his legions at the battle of Carrhae in May 53 B.C. With one stroke the position of the 
two antagonists changed, for the well-nigh invincible Romans were stunned by the 
great losses, and the Parthians became the champions of anti-Roman groups 
everywhere in the Near East, especially the Jews and some peoples of Anatolia. We 
are not concerned here with the change in Roman thinking about the east, but about 
Parthian reaction to the victory. The Parthians, contrary to what one might except, 
did not take advantage of the victory, for we hear only about raids into Syria with no 
thought of immediate conquest. Time was secured for the Romans who proceeded to 
consolidate their position in the east. It is from this period that the letters of Cicero, 
appointed governor of Cilicia in 51 B.C., give insights into the changed attitude of the 
Romans towards the Parthians. The Romans now realized they had a formidable foe, 
able and willing to challenge Roman domination of the east, although their cavalry 
tactics belied any intentions to permanently occupy settled areas. The need or desire to 
make a treaty between the two hostile powers, however, does not seem to have 
occurred to either side until the Roman civil war brought Pompey to seek an alliance 
with Orodes, the Parthian king (Cassius Dio, XLI, 55), but nothing happened since 
Julius Caesar triumphed over his rival. Caesar made extensive preparations for a war 

,0b Op. cil. [ch. 7, n. 64], 70-5, and in more 
detail, DobiaS, op. cil. (n. 40], 228^*2. 



234 Chapter VIII 

of revenge against the Parthians, but his murder ended such plans. 107 The conflict 
between the Republicans and Marc Antony and Octavian changed the picture, for 
Cassius, governor of Syria, took the side of the Republicans and sought support from 
Orodes (Cassius Dio, XL VIII, 24) and according to several sources Parthians fought 
against Octavian at the battle of Philippi in 42 B.C. 108 Antony took up the task of 
mobilizing for an expedition against Parthia, but while he was in Egypt the Parthians 
invaded Syria under Pacorus, son of Orodes, together with Labienus, the envoy of 
Cassius who had remained at the Parthian court. After initial successes they divided 
forces, Labienus moving into Asia Minor and Pacorus south into Palestine. Many 
local rulers broke with Rome and joined the Parthians and in Jerusalem Antigonus, 
nephew of Hyrcanus the high priest, was made king by the Parthians while Hyrcanus 
was carried a prisoner to Parthia. 109 Roman power reached its lowest ebb in the east, 
but soon the tide turned. 

Antony's general P. Ventidius Bassus in 39 B.C. defeated and killed Labienus, who 
had taken the title itnperator, and the following year Pacorus too met his death at the 
hands of the same general. 1 10 Antony came to Syria to restore Roman authority and 
in 37 B.C. Jerusalem fell; Antiogonus was killed and Herod was put in his place. 
Elsewhere pro-Parthian rulers were punished, and the Romans regained their former 
position. It is, of course, unknown whether Orodes had any dream or plan of re- 
establishing the Achaemenid Empire, but this was the only opportunity given to the 
Parthians in their history to extend their sway beyond the Euphrates to the 
Mediterranean Sea. Undoubtedly many of the inhabitants of Syria and Asia Minor 
welcomed the Parthians, but large numbers, perhaps the majority, wished to remain 
under Roman protection. In any case, the Parthians failed, and Orodes renounced his 
throne to his son Phraates IV, who then killed his father and brothers. Some Parthian 
nobles fled to the Romans, including a prominent general called Monaeses (mnyS) 
whom Antony hoped to use as a friendly claimant to the Parthian throne, but 
Phraates persuaded the deserter to return, and Antony seeing his chances diminished 
proposed a peace with Phraates, seeking the return of the standards lost by Crassus at 
Carrhae as well as war prisoners. Phraates refused, and Antony made preparations to 
attack Parthia through Armenia. 1 1 i The war between Antony and Phraates involved 
Armenia and Median Atropatene, and in this direction the Romans advanced as far as 
the capital called Phraaspa identified as modern Maragheh. 1 1 2 The king of Armenia, 

107 For an extensive survey of the sources on ius" (Sb. Akad. d. Wiss., MUnchen, 1974, Heft 1) 
Caesar's preparations see Debevoise, op. cit. [ch. 7, and "Marcus Antonius, Triumvir v. Herrscher d. 
n 64], 106-07. On the relations of this period see Orients" (MUnchen, 1977), 184 foil. For all of the 
Ziegler, op. cit. [n. 62], 32-34. wars see A. Giinther, Beitrage zur Geschidile der 

108 Debevoise, op. cit. [ch. 7, n. 64], 108, for Kriege zwischen Romern und Parthern (Berlin, 
sources. 1 922), 1 36 pp. Coins attributed to Pacorus suggest 

109 Ibid., 13. The events of this period are a coregency with his father. 

discussed in detail by A. G. Bokshchanin, Parjiya i ' ' ' On the attempts at peace see Ziegler, op. cit. 

Rim, 2 (Moscow, 1 966), 90-99. The overthrow of [n. 62], 36-44. 

the Hasmonaean house by the Idumaeans in ' ' 2 V. Minorsky, "Roman and Byzantine Cam- 
Jerusalem cannot be discussed here. paigns in Atropatene," BSOAS, 11 (1944), 258- 
1,0 On the Roman actions see H. Buchheim, 61, with references to sources. Perhaps this is a 
"Die Orientpolitik des Triumvirn M. Antonius," Middle Iranian form *frah-aspa, but the location is 
Abb. Heidelberger Akad. der Wiss., 3 (1960), 75-77 conjecture. 
L; H. Bengston. "Zum Partherfeldzug des Anton- 



The Parthians on the Plateau 235 

Artavazd, abandoned Antony and the ruler of Azerbaijan, also called Artavazd, who 
had been quarrelling with Phraates IV, became reconciled with the Parthians against 
the common foe. The result was a disastrous retreat of the Romans with great losses, 
but after the departure of Antony, Phraates and Artavazd, king of Azerbaijan, again 
quarreled, after which Artavazd then sought the support of the Romans. Plans were 
made for a new invasion of Parthian domains by Antony but instead he only entered 
Armenia and deposed the king in 34 B.C. and brought him and his family to 
Alexandria as captives. 

Antony entered the eastern game of marriage alliances to strengthen his position, 
offering his young son Alexander to the daughter of the king of Azerbaijan, after 
having first considered a marriage into the Armenian royal house. In general he 
distributed territories to dynasts who would support him, such as Cappadocia to 
Archelaus and Sophene to Polemo, according to Cassius Dio (XLIX, 32). The 
intermarriages between the minor courts under Roman control paralleled those 
under Parthian influence, and the picture resembles that of Europe in the nineteenth 
century, when the royal houses were filled with intermarriages. In 33 B.C. Antony 
and Artavazd of Azerbaijan defeated the Parthians, allied with Artaxias, the new ruler 
of Armenia and a son of the deposed king. But after Antony had to withdraw all 
Roman troops for his war with Octavian, his ally in turn was defeated and fled to the 
Romans (Cassius Dio, LI, 16). Roman power and influence again sank in areas to the 
east of the Euphrates River, but so volatile was the power structure in the east that 
Phraates was not able to take advantage of the Roman civil war between Antony and 
Octavian. Instead he was faced with a revolt led by a certain Tiridates who struck 
tetradrachms at Seleucia calling himself both 'friend of the Romans' and 
'philhellene.' 113 Even though he may have hoped for Roman aid to maintain his 
position, the victor at Actium did not support Tiridates, and he had to flee before the 
forces of Phraates, and took refuge in Syria in 26 B.C. (Cassius Dio, LI, 18). Octavian, 
now called Augustus, prepared for war, but then peace was made, and the Roman 
standards lost by Crassus and Antony were returned in 20 B.C. by Phraates, and the 
Romans celebrated this event by erecting a triumphal arch in Rome. 114 

Roman policy had vacillated between the creation of Roman provinces in the east 
to the installation or recognition of client kingdoms, but Parthia favored the latter 
course. Antony's policy of supporting pro-Roman dynasts was followed by Augustus, 
and he preferred intrigue and rewards and punishments with client dynasts rather 
than military action to extend Roman frontiers. In Armenia, for example, a pro- 
Roman party was probably encouraged to ask Augustus for a new king, and the 
future emperor Tiberius escorted Tigranes, the younger brother of Artaxias, to 
Armenia with Roman troops to install him on the throne. Fortunately for the 
Romans, Artaxias was murdered and Tigranes II was accepted by the Armenians and 
Romans thus averting a civil war. In an inscription at Ankara Augustus says that he 
made Armenia major into a province but 'he preferred to hand it over to Tigranes to 

1,3 Sellwood, op. cit. [ch. 7, n. 57], 167-68. See 64], 141, n. 58; he stresses the importance of this 

also H. Dessau, Prosoprographia Imperii Romani act for the Romans. On the treaty between the 

(Berlin, 1897-98), T 175. Romans and Parthians see Ziegler, op. cit. [n. 62], 

114 References in Debevoise, op. cit. [ch. 7, n 45-51. 



236 Chapter VIII 

rule.' 1 ' 5 About the same time the son of Artavazd of Azerbaijan, called Ariobarzanes, 
succeeded to the throne in that kingdom at the whim of Augustus according to his 
inscription (para. 27), but in reality Roman influence had little to do with this 
succession. Archelaus of Cappadocia was a Roman client, and after the settlement of 
the Armenian succession he, probably because of his cooperation with the Romans, 
received Armenia Minor, the land around Melitene (Malatya) from Augustus. 
Archelaus' wife was probably a princess of the Median royal family, an indication of 
the dynastic connections in this time. Armenia, although in the eyes of Augustus a 
client state, in reality was to prove a headache for the Romans, since pro-Parthian 
sentiments were strong, and the Parthians remained as influential as the Romans in 
influencing Armenian affairs. 

Augustus had sent a slave girl called Musa to the Parthian king, who managed to 
influence Phraates so much that he raised her from concubine to queen, and she 
secured the right of succession for her young son, born while she was a concubine, and 
the sending of other sons of Phraates to Rome for education, although the Romans 
considered them as hostages. 116 The Augustan peace gave a great impetus to 
merchants, and commerce flourished across the Euphrates, the boundary between the 
two states. The small trading states, such as Palmyra and Hatra, perhaps profited more 
than the two central great powers, but prosperity was manifest everywhere. The 
death of Tigranes II of Armenia some years before 6 B.C. almost interrupted the peace, 
for the Armenians raised the son of this Tigranes with the same name, and his sister- 
wife Erato to the throne without the approval, tacit or direct, of the Romans. 
Augustus, however, accepted the fact, but about 6 B.C. he apparently tried to impose 
another king on Armenia called Artavazd, possibly a brother of Tigranes III (Tacitus, 
Annals, II, 4). This move failed, since the pretender died, and Tigranes then did send 
gifts to Augustus to conciliate him (Cassius Dio, LV, 9), and the Romans accepted 
him. After Tigranes HI died fighting 'barbarians' about a.d. 2 (Cassius Dio, LV, 10a), 
his sister Erato could not hold the throne. If we follow the Res Gestae (para. 27), after 
the death of Tigranes III Augustus gave Armenia to Ariobarzanes, king of Media 
Atropatene (Azerbaijan), and after his death to his son Artavazd. He continues that 
when Artavazd was murdered after about two years of rule in a.d. 6 he sent Tigranes 
IV, a grandson of Herod the Great of Judaea by his son Alexander, to Armenia. Thus 
we see that the intervention of Augustus into the internal affairs of the small 
kingdoms of this area respected dynastic ties among them, as it generally ignored the 
wishes of the people. Augustus seemed to have followed a policy of strengthening the 
related dynasties, even though he may have thought he was dividing them. At the 
same time, unsuccessful rulers who fled to Roman protection were well received by 
Augustus, and in his Res Gestae (para. 32) he mentions Artaxares of Adiabene, as well 
as Artavazd of Media and others, who took refuge as suppliants of the Roman 
emperor. A similar policy was followed by the Parthians, although we do not have 
the sources in this regard. 

115 T. Mommsen, Res Gestae Divi Augusli " 6 Ziegler, op. c it. [n. 62], 52, and Debevoise, op. 

(Berlin, 1883). par. 27. Tindates II should really be cit [ch. 7, n. 64], 144, for sources. 
Ill, but Tindates the 'Great' is usually called I, so we 
use II here. 



The Parthians on the Plateau 237 

Phraates IV died, possibly poisoned, in 2 B.C. and was succeeded by his widow 
Musa who married her son Phraataces, and issued a joint coinage, the only such 
example in Parthian history. The Parthians did not support Phraataces, however, and 
in a.d. 4 he fled to Syria where he vanishes from the scene. 1 ,7 He was followed by 
Orodes III who struck tetradrachms alone and had a very short reign and was 
assassinated. The Parthian nobles sent envoys to Rome to obtain another son of 
Phraates IV, who had been sent there, to be their ruler, and Vonones the eldest was 
sent home in a.d. 8. His Roman ways did not appeal to the Parthians, and a claimant 
to the throne, Artabanus from Hyrcania, led a revolt against Vonones, who at first 
was successful and struck coins with the legend 'conqueror of Artabanus,' but 
Artabanus prevailed and drove Vonones out of the land. Vonones fled to Armenia 
where Tigranes IV had been deposed, and the throne was vacant. For little more than 
a year Vonones occupied the throne of Armenia, until Artabanus threatened an 
invasion and Vonones left for Roman exile in a.d. 16 (Tacitus, Annals, II, 1-4). The 
Romans, however, were not ready to abandon Armenia to Orodes son of Artabanus, 
so the new emperor Tiberius sent his adopted son Germanicus in a.d. 18 with an 
army to install Zeno, son of Polemo king of Pontus, as king of Armenia, with the new 
name Artaxias, and from 1 8 to 24 peace prevailed in Armenia, as well as between 
Rome and Parthia which gave Artabanus the opportunity to consolidate his rule. 118 

Our sources give little information about internal affairs in Parthia, but the coinage 
of Artabanus may reflect a turning point in Parthian history, for the epithet 
'philhellene' which had appeared on the issues of previous kings was omitted on some 
issues, presumably late tetradrachms. This could signify an emphasis on an Iranian 
reaction to the Romanized Vonones whom Artabanus defeated, but it would be 
wrong to conclude solely from the coins that Artabanus followed an anti-Hellenic 
movement, since later coins return to the epithet which persisted down to the end of 
the dynasty. Artabanus, however, was from a collateral branch of the Arsacid family 
and thus represents a change from a pro-Roman to an independent Iranian policy for 
the Parthians. Josephus (Antiquities, XVIII, 48) says Artabanus was king of Media, 
while Tacitus (Annals, II, 3) says he grew up among the Dahi, to the east of the 
Caspian Sea where his family had marriage connections. Kahrstedt goes to great 
lengths to prove that Artabanus was not king of Atropatene but of Hyrcania. 119 
Certainly Artabanus, according to our sources, took refuge and secured support in the 
east when pressed by Vonones and later rebels, but he also could have held kingship in 

117 Ace. to the Res Gestae (par. 32), he took years in the period a.d. 15-18 and perhaps fled to 
refuge with the Romans. See also Josephus, Jewish Parthia where he reappears at the head of a 
Antiquities, XVIII, 42, where he attributes the Parthian army to avenge the murder of his 
uprising of Parthians to revulsion over the brother. See P. Asdourian, D/V politischen Beziehun- 
mother-son marriage. That Musa allowed herself gen zwischen Armenien und Rom (Venice, 1911), 
to be deified is suggested by a ring from Anatolia 81. This is uncertain, however, and conjectural, 
where she is identified as Thea Urania, perhaps to ' 19 U. Kahrstedt, Artabanos 111 und seine Erben 
be identified with Anahita in Iran. Cf. R. Zahn, (Bern, 1950), 12-16. In his arguments for Hyr- 
"Ein kleines historisches Monument," Anatolian cania as the homeland of Artabanus he is 
Studies presented to Wm. Ramsay, ed. by W. H. convincing but not so in including Carmania or 
Buckler (Manchester, 1923), 454-55. greater Kerman as the other area of support for 

118 The history of Armenia is confused in this Artabanus as a dual kingdom, 
period. Orodes may have ruled Armenia for a few 



238 Chapter VIII 

Atropatene for a short period, since the local dynasty seems to have either died out or 
was replaced. In any case, Artabanus after defeating Vonones put his eldest son on the 
throne of Atropatene, which bound it closer to Parthia. The period from the accession 
of Artabanus to the end of Parthian rule is one of conflict followed by conciliation 
and then new hostilities with Rome. As many modern scholars, as well as ancient 
authors, have remarked, the Romans now considered the Parthians worthy 
opponents, and treaties between the two parties in 20 B.C., in a.d. 1 between Caius 
Caesar, grandson of Augustus, and Phraataces, in a.d. 18 or 19 between Artabanus 
and Germanicus, nephew of Tiberius, and in a.d. 37 between Artabanus and L. 
Vitellius, Roman governor of Syria, all testify to the political equality of the two 
opponents. 120 Most of the reign of Artabanus was spent in restoring prestige and 
authority to the central government, and on the whole he was successful. 

In Armenia, however, at the death of Zeno-Artaxias in 34, Artabanus installed his 
son Arsaces as king, and Tiberius sent a counter-claimant Tiridates, of the family of 
Phraates. Tiridates shortly died, and the Romans turned to Mithradates, brother of 
Pharasmanes, king of Iberia (Georgia), persuading him to seize the Armenian throne 
after the murder of Arsaces. Mithradates was successful and defeated the Parthian 
force led by another son of Artabanus, Orodes, who was killed. Mithradates then 
ruled Armenia amidst constant intrigues and warfare until 47. Artabanus at this time 
lost his influence, and a revolt of nobles caused Artabanus to flee to the east, as he had 
done in the time of Vonones (Tacitus, Annals, VI, 36). Again the Romans sent a 
pretender to the Parthian throne, Tiridates grandson of Phraates IV, across the 
Euphrates where he was crowned king. But Artabanus was able to return and 
Tiridates fled to Roman territory, and the peace of a.d. 37 was the result. This did not 
end internal problems for Artabanus, and the rest of his reign was troubled by the 
unrest of the nobility such that on one occasion Artabanus had to take refuge with 
Izates II of Adiabene, a client king. At the death of Artabanus, central power and 
authority had been shattered, and constant bickering for the throne followed him. 
The weakness of central authority which even an energetic ruler like Artabanus could 
not overcome became the hallmark of later Parthian history. The flights of Artabanus 
to Hyrcania and his restorations cannot have contributed to the stability of the state, 
and territory, or at least jurisdiction over some areas, was lost in the east as well as the 
west. The district of Herat may have been lost to Gondophares in the reign of 
Artabanus, for his coins predominate among the Parthian coins overstruck by 
Gondophares. 121 At the same time that Parthian central authority suffered, Parthian 
princes were installed as client rulers in various areas and thereby Parthian influence 
was spread through intermarriage more than conquest. But the client princes had 
equal claim to the Parthian throne, and civil wars became endemic. 

120 On the treaties cf. Ziegler, op. cit.[n. 62], 48- gases, who revolted against Artabanus and sup- 
64. Most of the provisions of the agreements ported Tiridates, with the Indo-Parthian ruler of 
stressed the return to a status quo after hostilities, the same name is most unlikely. Also his statement 
and the Euphrates border of the Romans remained that under Vologeses I Persis was lost by the 
constant, whereas Armenia continued to be an area Parthians makes little sense; Persis continued as a 
of discord. client state to the end of the dynasty, even though 

121 Kahrstedt, op. cit. [n. 119], 34-35. All of in some periods Parthian influence was stronger 
Kahrstedt's surmises, given as history, need not be than in others. The striking of coins is no sure sign 
accepted; for example, the identification of Abda- of absolute independence. 



The Parthians on the Plateau 239 

At the death of Artabanus, presumably his son Vardanes succeeded him, although 
his descent is uncertain as is that of his rival Vologeses I. Coins fail us, primarily 
because the city of Seleucia, with dated tetradrachms, had revolted in the last years of 
Artabanus, and we do not know when Vardanes or Gotarzes ruled there. 122 Civil 
war, interspersed with truces, lasted until 47, when Vardanes was murdered, and 
Gotarzes became sole ruler. Some of the Parthian nobles, headed by a member of the 
Karen family, opposed Gotarzes and turned to Rome, asking that Mithradates (called 
Meherdates, the current pronunciation, by Tacitus), son of Vonones and grandson of 
Phraates, be sent to be king. He was defeated and captured by Gotarzes, however, and 
the outside threat ended. On the death of Gotarzes in 51 by disease or accident, he was 
succeeded by Vonones, who had been ruler of Atropatene, according to Tacitus 
(Annals, XII, 14, 7), but who ruled a very short time and was followed by his son, 
Vologeses 1. His brother Pacorus became king of Atropatene, and he determined to 
put another brother Tiridates on the throne of Armenia, which had become vacant by 
the murder of Mithradates in 52. Vologeses was successful, but Rome was aroused, 
and the new emperor Nero, who succeeded Claudius in 54, sent an experienced 
general Co. Corbulo to retrieve Roman prestige; for the ebb and flow of Roman- 
Parthian relations, especially in regard to Armenia, mostly concerned a matter of 
honor for the respective parties rather than actual gain or loss of territory, since 
neither side was able to incorporate Armenia as an integral part of their own domains. 
The struggle on the frontiers of the Roman and Parthian empires was to continue 
through Byzantine and Sasanian times to the coming of Islam, a period of more than 
half a millennium, and the two arms of the 'Fertile Crescent,' Syria and Iraq, although 
united in language and culture, were to develop different traditions, one looking to 
the west and the other to the east. In Classical sources we find references to the desire 
of the Iranians, first Parthians then Sasanians, to reach the Mediterranean and restore 
the empire of the Achaemenids, even though any memory of the Achaemenids was 
becoming hazy for the Iranians. 

The war between Vologeses and Corbulo was complicated for the Parthians by 
revolts in the east, according to Tacitus (Annals, XIII, 7 and 37) by a son of Vardanes 
and the Hyrcanians. This latter revolt was serious, and ambassadors from the 
Hyrcanians came to Corbulo, and the revolt seems to have resulted in a kind of 
independence for that eastern region from direct Parthian control (XV, 2). In 58 
Corbulo, however, drove Tiridates from Armenia, although hostilities did not end, 
for Tiridates continued to struggle, supported by his brothers Vologeses of Parthia 
and Pacorus of Atropatene. The Romans, however, installed Tigranes, great-grandson 
of Archelaus of Commagene, as their king of Armenia, and Tigranes in 61 launched 
his own invasion of Adiabene which provoked a countermove by the Parthians. 
After much fighting, a peace was arranged in 63, and the conflict came to an end with 
the understanding that Tiridates would be recognized as king of Armenia, but that he 
would receive a crown from Nero in Rome, an indication of a nominal Roman 

122 Josephus, Antiquities, XX, 3, 4, implies that Gotarzes, as Josephus says, then the struggle 

Vardanes immediately succeeded Artabanus, his between Vardanes and Gotarzes could have been 

father, while Tacitus, Annals, XI, 8, suggests that an attempt of the Arsacid ruler of Atropatene to 

Gotarzes succeeded Artabanus on the throne. If take over the central power. 
Vologeses, from Atropatene, was a brother of 



240 Chapter VIII 

hegemony over Armenia, but in reality it meant little. The trip of Tiridates to Rome 
and the celebrations which took place there in 66 were recorded by Dio Cassius 
(LXII) and Tacitus (XVI, 23), and they heralded a peace between the Romans and 
Parthians which lasted for half a century. 123 Rome had failed to impose its will, 
perhaps not understanding fully the importance of local loyalties to the intermarried 
royal houses of the principalities in this part of the Near East. Perhaps Roman supply 
lines and difficult logistics made Roman attempts either to make Armenia a province 
of their empire or a client state with a Roman-appointed ruler unfeasible, but, in any 
case, Corbulo was the agent of the change in Roman policy. The marriage ties of 
Parthian and Armenian nobility certainly did not help the Roman cause in Armenia 
or in other frontier states. The later Roman answer to Parthian inter-family 
connections in the east, the conferring of Roman titles of general, senator or consul on 
local dynasts, also failed to win support, and the Arsacid family connections in the 
courts of Armenia, Adiabene and others were to prove more important. 

From notices in Tacitus and Cassius Dio about Hyrcanian embassies to the Romans, 
modern authors have deduced the existence of a Roman client state in the east. 124 
More likely is simply the opposition of local inhabitants to demands of the Parthian 
king and his refusal to recognize the rights of Hyrcania similar to Atropatene or 
Armenia. Without information, it is surely excessive to call Hyrcania a Roman client 
state because the Hyrcanians sought Roman help. Based on geographical data from 
Ptolemy and elsewhere, Schur (op. cit., 64-79) constructs a history of two powerful 
states in eastern Iran, Aria and Hyrcania, which expanded and contracted according 
to Parthian involvement with Rome. All is conjecture and must be so regarded. The 
Parthians themselves sought Roman aid against an invasion of the Alans from north 
of the Caucasus in 75 but it was refused by Vespasian which soured the Roman- 
Parthian friendship. 1 25 It is not possible to discuss Roman designs for the conquest of 
the Albanians, and the desire to open a land route to India over the Caspian through 
Hyrcania, if the Romans really had such extensive plans, but the Parthians in this 
period were hardly as isolated and reduced to a small area of rule as some scholars have 
proposed. 126 Vologeses rather increased his authority and prestige compared to his 
predecessors; he founded a city Vologesia as a rival to Seleucia and was interested in 
promoting trade. 127 He has been characterized as an anti-Hellenic king who 
promoted an Iranian cultural reaction solely on the basis of Aramaic letters appearing 
upon a few of the coins he struck, which is hardly strong evidence for a reversal of 
policy, since his coins continue with the legend 'philhellene.' Likewise the possible 

123 For details of this period see W. Schur, Die l25 Ziegler, op. cit. [n. 62], 80. 
Orienlpolilik des Kaisers Nero, Klio Beiheft, 15 '"Schur, op. cit. [n. 123], 80-85. Romans did 
(Wiesbaden, 1923), 29-32. Schur does not sub- reach the shores of the Caspian as a Latin 
scribe to the theory that Tigranes was put aside by inscription of the XII legio Fulminata from the 
the Romans because he could not command time of Domitian (81-96) indicates. See K.Trever, 
support among the Armenians, but this seems to Ocherki po istorii i kulture Kavkazskoi Albanii 
have been the case, for the Roman or Augustan (Moscow, 1959), 342-46. The significance of this 
policy of imposing a Roman vassal in Armenia was inscription, however, should not be inflated, for 
a failure. Cf. Ziegler, op. cit. [n. 62], 75-77; conquests of Domitian so far east are unknown in 
Kahrstedt, op. cit. [n. 119], 83, and others. literary sources. 

124 Schur, op. cit. [n. 123], 37-38; cf.Josephus, 127 A. Mariq, "Vologesias, Temporium de Ctesi- 
Wars, VII, 245. phon," Syria, 36 (1959), 271. See n. 91. 



The Parthians on the Plateau 241 

attribution to this king of the collecting of the fragments of the Avesta in the later 
Middle Persian book the Denkart also does not mean such a change. The Romans 
evidently had a great respect for Vologeses, since they expanded the system of roads 
and fortifications in Syria under the Flavian emperors who succeeded Nero. 128 

The end of the reign of Vologeses, however, is unclear, for the coins seem to 
incidate a conflict with a Pacorus, whose relationship to Vologeses is unknown, but 
the latter's coins end about 78 or 79, while coins of Pacorus begin about the same 
time. 129 An unresolved numismatic problem is the existence of coins with the name 
Vologeses but with a completely different bust and crown than the usual issues 
of Vologeses. Some numismatists have postulated another Vologeses, a rebel against 
Vologeses I, while others have attributed the coins to a later king with the name 
Vologeses. 130 A later Byzantine source, Zonaras (XI, 18 or 578 C), mentions a 
Parthian king Artabanus as ruling Parthia at the time of the eruption of Mt. Vesuvius 
in 79, but he is mentioned nowhere else, although coins have been attributed to him. 
One incident mentioned by the sources is that of the pseudo-Nero, a Roman who 
claimed to be the murdered emperor, but who had to take refuge with the Parthians. 
This did not serve to improve relations between the two powers, and continued 
fortification of Roman border territory, the limes, did not augur well for future 
peaceful relations between the two. The balance between the powers was to be upset 
by Trajan who became Roman emperor in 98. 

In Parthia Pacorus had a long reign not free from trouble, however, of which we 
have no information, but the long series of coins indicate a long rule, perhaps to 105, 
although the end date of his rule is unknown. Internal affairs in the Parthian domains 
are veiled in this period, but at the time of Trajan's accession in 98, hints in Classical 
sources indicate internal instability and perhaps even civil war there. A certain 
Chosroes (hwsrw), perhaps a brother-in-law of Pacorus, issued coins and at the same 
time so did Vologeses II, whose relationship with Pacorus is unknown. 131 We may 
only say that the reigns of Chosroes (also called Osroes on coins) and Vologeses were 
contemporaneous. The pedigrees and the reigns of later Parthian kings are uncertain, 
and the entire second century is a 'dark period' of Parthian history since there are no 
inscriptions, the coins are highly stylized, and the Romans only showed an interest 
when they invaded Parthian domains. This is why we cannot tell whether there were 
two rulers called Vologeses in this period, based solely on coins. 

In the time of Trajan, we hear of a Parthian general Sinatruces, son of Mithradates 
and father of a Vologeses who received a portion of Armenia to rule from the 
Romans, according to Cassius Dio (LXVIII, 30), but whether this Sanatruces is to be 

1 28 Ziegler, op. cit. [n. 62], 80, for references. and frequently the personal name of the king. The 

129 Sellwood, op. cit. [ch. 7, n. 57], 220-26, drachms, which are much more numerous, are 

proposes two kings of the same name, Vologeses I highly stylized and with several exceptions give no 

(51-78) and Vologeses II (77-80), on the basis of personal names. Furthermore, local kings in 

different coin types, but this is hardly enough to Atropatene and elsewhere on the Iranian plateau 

prove a second ruler at that time. may have struck coins in Parthian style, so 

130 Ibid., 226-28; Debevoise, op. cit. [ch. 7, n. numismatics must be used with care. For a new 
64], 214, n. 3. method of identification on the coins see D. G. 

131 In relying on numismatics it should be Sellwood "A Die-Engraver Sequence for Later 
emphasized that only Parthian tetradrachms, Parthian Drachms," NC, 1 series, 7 (1967), 13-28. 
which were minted at Seleucia alone, give a date 



242 Chapter VIII 

identified with a ruler of the same name in Edessa, c. 91-109 and whether this 
Vologeses later became Vologeses II (or III) of Parthia is mere conjecture. Names of 
Arsacid princes do occur in the Classical sources, but it is not possible to make a 
genealogical table of the Arsacids in this period. 

The war of Trajan against the Parthians has been studied by many scholars, notably 
by J. Guey followed by Lepper whose reconstructions generally have been 
accepted. 132 We are not here concerned with the causes of Rome's aggression, 
whether economic to control trade routes to the east, or a personal desire of Trajan for 
fame and glory, or, as suggested only by Lepper, as an attempt to stabilize the frontier 
by advances into enemy country from the limes which Trajan had continued to build, 
following the Flavian emperors. Certainly Rome was the aggressor, even though 
Armenian affairs became a pretext. For Chosroes encouraged one son of Pacorus, 
Parthamasiris, to replace his brother or half-brother called Axidares in Armenia 
which happened, but since the latter had the support of Rome, Trajan arrived in the 
east in 1 14 to begin a war against the Armenians in which he was successful such that 
Armenia was proclaimed a Roman province. Trajan organized better the limes 
system, and in 116 he invaded Adiabene and put its king Mebarsapes to flight. The 
land of Adiabene was annexed as well as the entire basin of the Tigris-Euphrates as 
Maricq has brilliantly shown. 1 33 The Romans, however, were far too extended, even 
if there was no unified Parthia opposing them, and revolts broke out in 116, and 
Trajan was obliged to retreat from Ctesiphon which he had captured in 1 1 7. He failed 
in his attempt to capture Hatra, the caravan city in the desert of Mesopotamia, and 
shortly afterward he died of illness. We are not here concerned with Roman history, 
but the farthest advance of the Romans, to the Persian Gulf, under Trajan must have 
made a strong impression on the Parthians. Some scholars, such as Maricq, have 
argued that Trajan intended to advance the boundaries of the Roman Empire to the 
Zagros mountains in the east, a natural barrier. This view, however, conflicts with the 
activity of Trajan in building roads and forts along the limes of the Syrian desert and 
upper Mesopotamia, and Lepper's view that Trajan in reality followed a policy of 
penetration beyond the limes to secure the real borders of the empire, behind the limes, 
seems more accurate. Trade with the east was surely important, but any plan to 
incorporate all the land to the Persian Gulf must have seemed unrealistic to many 
Romans, as it did to Hadrian, successor of Trajan. Perhaps the most important result of 
the peace of Hadrian was the abandonment of the Trajanic policy of annexation of 
client states in the east as provinces of the Roman Empire, and a return to the policy of 
client kingdoms. This gave an opportunity for small trading city-states such as Hatra, 
Mesene and Palmyra to flourish. 

132 J. Guey, "Essai sur la guerre parthique de l3i A. Maricq, "La province d'Assyrie' cree par 

Trajan," BiblhtMque J'htros, 2 (Bucharest, 1937), Trajan," Syria, 36 (1959), 257. Whether most 

160 pp., and F. A. Lepper, Trajan's Parthian War Romans thought they could hold the three new 

(Oxford, 1948), 224 pp., with further references. provinces of Armenia, Mesopotamia and Assyria 

Lepper systematically analyzes especially the chro- (Babylonia) with the client state Mesene, at the 

nology of Trajan's campaigns. E.J. Keall, "Osroes : head of the Persian Gulf, is conjectural. Trajan 

Rebel King or Royal Delegate?" Cornucopiae, 3 must have considered his conquests as permanent, 

(Toronto, 1975), 17-32, argues that Chosroes while his trip to the Persian Gulf had something of 

never acted in concert with Trajan but always in the bravado of Alexander the Great, 
defense of his suzerain Vologeses. 



The Parthians on the Plateau 243 

As noted, the second century is a dark century in Parthian history, and the reigns of 
Hadrian, Antoninus Pius and Vologeses II reveal no activity on the Parthian frontier. 
The mention of Bactrian and Hyrcanian envoys in the time of Antoninus indicates 
the continued independence of Hyrcania from the Parthian central authority while 
the Bactrians are surely the Kushans. 134 Vologeses III (or IV; c. 147-191) may have 
taken advantage of the accession of Marcus Aurelius to break the long peace and the 
fixed frontiers of the Euphrates between the two states, but the apparent reason for the 
war was his attempt to dislodge a Roman client ruler in Armenia who had been 
installed by Antoninus about 140. 135 The Parthians were successful in putting a new 
king on the throne of Armenia, called Pacorus, and in annihilating a Roman army led 
by the governor of Cappadocia, Severianus. The Romans, however, retaliated, and a 
strong army soon took the Armenian capital of Artaxata in 163 and replaced Pacorus 
with a prince, Sohaemus, of the royal family of Edessa and a Roman senator as well. 
To the south another army in 164-165 advanced as far as Seleucia which was 
destroyed by the Roman general Avidius Cassius, while Ctesiphon, the capital, was 
taken and plundered. Sickness, however, caused a retreat of the Romans, and peace 
was reestablished in about 166. Under the peace treaty, which is not mentioned by 
our sources, the old boundaries were rectified a little in favor of the Romans, by 
making the town of Nisibis and the Khabur River with the Singara mountains (Jibal 
Sinjar) the boundaries of Roman territory. Armenia's Sohaemus was recognized as 
king. Peace remained between the two states even when the opportunity to support 
Avidius Cassius in his revolt against Marcus Aurelius in 175 was presented to 
Vologeses. The quick fall of Cassius and the intervention of the Roman emperor in 
the east to regulate affairs with local rulers, and with ambassadors from Vologeses, 
strengthened the peace which lasted through the reign of Commodus (180-192). 

In Parthia Vologeses III seems to have had a rival called Chosroes who is known 
only by his coins, and he cannot have reigned long even over only a part of Parthia 
towards the end of the rule of Vologeses. 136 The latter was succeeded by another 
Vologeses whose relationship to his namesake is unknown. The next to last Vologeses 
supported Pescennius Niger as claimant to the Roman throne in 193, but the victor 
was Septimius Severus. The Parthians and their allies, however, in the period of 
Roman civil war, had taken some territory and towns and refused to return them to 
Roman rule. 137 Thus Severus crossed the Euphrates in 196 and had some success, but 
in 196 he was recalled to Gaul by a revolt. Nonetheless, the Romans maintained their 
eastern boundary of the Khabur and Singara mountains. The absence of Severus 
emboldened the Parthians to attack, and much territory in Mesopotamia came into 
their hands. Severus, having settled affairs in the west, returned and invaded the 
Parthian domains, capturing and sacking Ctesiphon in 198, and Vologeses fled from 
the city. Again it was not the Parthians but rather the devastated countryside which 

134 Envoys came in the time of Hadrian, l36 Sellwood, Coinage of Parthia [ch. 7, n. 57], 
according to the Scriplores Hisloriae Augustae, 281. No mention of this Chosroes is found in any 
Hadrian, 21, 14, and in the time of Antoninus literary source. 

(Aurelius Victor, De Caesaribus, Epitome 15, 4). 137 By allies, the rulers of Adiabene and 

135 For sources see M.-L. Chaumont, Recherches Osrhoene (to the west of Nisibis) are meant. Cf. 
sur I'histoire d'Armtnie (Paris, 1 969), 1 5-1 6. Arme- Cassius Dio (LXX V, 1 ). 

nian history in this period is also dark. 



244 Chapter VIII 

caused a Roman retreat, this time up the Tigris River. A long siege of Hatra in 199 
failed, and Severus had to return to Syria and apparently peace was made, based on the 
status quo before the war, although sources again do not tell us about a peace treaty. 
Vologeses died about 207 and was succeeded by his son of the same name, but 
sometime later Artabanus, another son, contested the throne; the civil war was 
incited by Caracalla according to Cassius Dio (LXXVIII, 2a). Caracalla looked for 
fame in conquest of the east, and in 214 he found a pretext in two exiles who had 
taken refuge with the Parthians, but Vologeses, pressed by internal problems, 
surrendered the fugitives to Caracalla. The latter, however, was determined to find an 
excuse for invasion, so when Artabanus gained the upper hand in Parthia in 216 the 
Roman emperor asked for the hand of the daughter of Artabanus in marriage which 
was refused by the Parthian king. Caracalla then invaded Adiabene, and Artabanus 
fled to the east but soon returned to the attack. Caracalla was assassinated, however, in 
217, and he was succeeded by Macrinus who sought peace. Artabanus rejected the 
overture for peace by the Romans and advanced toward Nisibis where an indecisive 
battle was fought, after which peace was made by the payment of an indemnity by the 
Romans to the Parthians. The end of the Parthian Empire, however, was in sight, and 
only the final dates of the two Parthian rulers, Vologeses and Artabanus, are in doubt 
because of coins which indicate that the former continued to rule until 228 while 
Artabanus continued to 227. 138 The Parthians were to fall to a new dynasty from 
Persis, the Sasanians. 

HELLENISM AND THE IRANIAN REVIVAL 

The Parthians have long suffered denigration from their successors, the Sasanians, as 
well as from their enemies, the Romans, and modern scholars usually have followed 
the ancient sources to give bad publicity for the Parthians as destroyers of the Hellenic 
heritage of the Seleucids. This reputation is undeserved, for the Parthians were neither 
enemies and destroyers of Hellenism nor traitors to the Iranian heritage of the 
Achaemenids and non-Zoroastrians, as has been asserted. For almost half a 
millennium the Parthians dominated the history of the Iranian plateau, and they 
finally were recognized as worthy opponents and equals in warfare and diplomacy by 
the Romans. 139 A review of the cultural achievements of the Parthians is in order. 
To begin with Hellenism, it must be re-emphasized that the epithet 'philhellene 
remained on most of the coins struck by the Parthian rulers to the end of the dynasty, 
and there is no evidence of either a prolonged or effective policy of attack on Hellenic 
culture of any of the Arsacid kings. The tradition of independence of those cities 
called polis was also continued from Seleucid times through most of the Parthian rule. 
The two most striking examples, of which we have source material, are Seleucia and 
Susa, both of which issued their own coinage and maintained their own institutions 
from Seleucid to Parthian rule. The Greek influence in both, quite naturally, became 

U8 See B Simonetta, "Vologese V, Artabano V son of Artabanus called Artavazd(es) (on the coins 

e Artavasde." NwHiiSHian'™, 19-20 (Perugia, 1953- read \tbnw for \twzd) who, according to 

54), 1-4. Sellwood, op. cit. [ch. 7, n. 57], dates the Simonetta, ruled after 227 for a short time, 
coins of Vologeses to 228 but Artabanus only to ' 39 See the convincing arguments of Ziegler, op 

224, while he correctly denies the existence of the cit. [n. 62], 140, el passim. 



The Partitions on the Plateau 245 

weaker, as the Hellenic population became absorbed by local people, but Hellenic 
features nonetheless persisted in these two cities. In December of the year 21 A.D. 
King Artabanus III wrote a letter to the city of Susa relating to the election of a certain 
Hestiaios, and, as has been remarked, the letter which is preserved on an inscription is 
to a Greek polis, of which the constitution is Greek, and the administration of the city 
is Greek. 140 Even though an Arsacid era beginning 247 B.C. was introduced by the 
Parthians, the Seleucid calendar remained more popular and, just as the native name 
Susa was used for the city, so also the Seleucid designation Seleucia on the Eulaius 
River remained in use until the end of the Parthian rule, all evidence of the tolerance 
for and even support of Hellenism by the Parthian rulers. Shortly after Artabanus, c. 
45, the kings of Elymais took the city, and they issued coins in their form of the 
Aramaic alphabet and a dialect of the Aramaic language. The rulers of Elymais 
remained in control of Susa until the end of the Parthian dynasty. An inscription of 
the last Artabanus, in the Parthian language and alphabet and dated to 215, found in 
the excavations, attests a return of the city to Arsacid allegiance. 141 Seleucia on the 
Tigris did revolt against the Parthians, but it was destroyed by the Romans. In both 
cities royal and local coinage existed, and it seems as though the Parthians were just as 
much, if not more, champions of Greek culture as the Romans. In any case, as far as we 
can see, Hellenism was not proscribed under the Parthians. 

One subject which has not been touched is commerce and trade, but we have no 
information from Parthia proper about such matters; rather the small states of the 
'Fertile Crescent' and even the Kushans in the east supply us with some information 
even though sparse. The Parthians were not great traders or merchants, but they did 
not by any means follow an anti-commercial policy but rather the contrary. One 
factor in the decline of Parthia in its last two centuries can be seen in the debasement of 
the coinage beginning with Artabanus III. Compared to earlier coins not only do the 
style and quality of the coins suffer, but in the tetradrachms the amount of silver 
declines. 142 Whether this is the result of the loss of silver mines or, more likely 
economic crises affecting especially the mint of Seleucia, we do not know. The 
population and cultivated land of the Susa plain in Parthian times was about three 
times that of the Achaemenid period according to an archaeological survey. 143 In the 
Parthian period land under irrigation and cultivation was increased over earlier times 
but not so much as the maximum use under the Sasanians. Again our sources fail us. 
The Roman wars undoubtedly adversely affected trade and commerce in the Parthian 
realm, but central weakness was more important than other factors contributing to 
the fall of the Parthians. 

The art of the Parthians has been discussed many times, and it is now generally 
agreed thatjust as the Parthians did not impose themselves on local rulers and cultures, 

140 Le Rider, op. cit. [ch. 6, n. 25], 35-36 and the tetradrachms, see W. Wroth, Catalogue of the 

421-22. Coins of Parthia, in The Greek Coins in the British 

'■" R. Ghirshman, "Une bas-relief d'Artaban V Museum, (London, 1903), lxv, and Sellwood, op. 

avec inscription en pehlevi arsacide," Monuments cit. [ch 7, n. 57], 5. 

Piot, 44 (1950), 97-107, corrected by W. B. u3 R. J. Wenke, "Imperial Investments and 

Henmng in "The Monuments and Inscriptions of Agricultural Development in Parthian and Sasan- 

Tang-i Sarvak," /1M, 2 (1952), 151-78. lan Khuzestan," Mesopotamia, 10-11 (Florence, 

' 42 On the debasement of the coinage, especially 1 975-76), 43. 



246 Chapter VIII 

so in the arts they allowed local schools to flourish. The Arsacid kings were not only 
'philhellene' in policy towards Hellenized conquered peoples, but also in the arts, but 
this early dependence changed in the last two centuries of Parthian rule and we find a 
'Parthian' style developing. In painting and sculpture the concept of 'frontality' 
dominated Parthian art after the beginning of our era. Without discussing many 
questions about the origins of this style and other features of Parthian art, it should be 
noted that Parthian art and architecture, such as is preserved, both are 'popular' in the 
sense that both Achaemenid and Sasanian art are not. 144 Rather in them the 
overwhelming stamp of the rulers is obvious. The Parthian age was not an 'imperial 
age' as both Achaemenid and Sasanian, but Parthian remains reflect rather the many 
currents of culture among the populace. The more that archaeologists uncover from 
Parthian sites, the more significant appears the importance of the Parthian period as a 
prelude to Sasanian art, culture and institutions. The opposition of Hellenistic to the 
'Oriental' art of the Parthians has been overly stressed, in my opinion, as has been the 
dichotomy between 'East and West' represented by Parthians and Romans. This is not 
to deny the fundamental differences between 'theoretical' Hellenistic and Roman art 
as opposed to 'theoretical' Parthian art, between representation of nature or realism as 
opposed to expressionism and stylized art, so ably sketched by Avi-Yonah. 145 The 
situation, however, was complex, and the influences of so many peoples such as 
Armenians, Nabataeans, Mesopotamians and others make the cultural and artistic 
panorama of the Near East at this time more complicated than the 'Hellenistic- 
Oriental' division. Perhaps one should look at the last century and a half of Parthian 
rule and from the meager sources show the change from an earlier 'Hellenistic' 
dominated age to one of Parthian autonomy in the realm of culture and institutions. 
In art the Hellenistic heritage had changed in the first century of our era from a 
syncretic Hellenistic-Iranian koint to the Greco-Buddhist or Gandharan art in the east 
under the Kushans, and in western Iran to a Parthian art with total 'frontality,' 
portrayal of the 'Parthian gallop' with horsemen in paintings or sculptures, the 
Parthian costumes and the use of ayvans in architecture and domed vaults, all 
hallmarks of later Parthian culture. In writing, Greek had lost its predominance, and 
Aramaic had been replaced by Parthian, a change symbolic of the change from early 
to late Parthian times. 146 

144 On 'frontality' see G. A. Koshelenko, "O 2 (1963), 69-71. In building techniques a differ- 
frontalnosti v Parfyanskom lskusstve," in Istori- ence between Greeks and Parthians could be 
arkheolog. sbornik v chest A. V. Artsikovskogo illustrated in the use of pillars; for the former they 
(Moscow, 1962), 135-36, who stresses the reli- were fundamental and walls were added to them; 
gious-ideological changes which induced artists to for the Parthians walls came first and pillars were 
stress frontality. Schlumberger, L'Oriem Hellinisi decorations. 

[ch. 7, n. 76], 198, stresses the origin from archaic 146 On Aramaic to Parthian see the discussion in 

Greek art, while M. A. R. Colledge, Parthian P W.Coxon, The Nisa Ostraca: Ur- Ideographic 

Art (Ithaca, N.Y., 1977), 143^*4 summarizes Texts?" AAH, 21 (1973), 185-204, and the 

'frontality' in this period. On the ancient Oriental examples of late Parthian writing on inscriptions, 

origin of 'frontality' see M. Avi-Yonah, Oriental p. 56. 'rtbnw MLKyn MLK ' BRY ivlgiy MLKyn 

Art in Roman Palestine (Rome, 1961), 76-79. MLK'BNYt hnsk ZK ZY hwsk Swi hitrp "Arta- 

145 Avi-Yonah, ibid., 10-12. It is not possible to banus King of Kings, son of Vologeses, King of 
discuss such fascinating topics as round cities of the Kings built this 'stele' which is of h"'sk, satrap of 
Parthians and the architectural innovations which Susa," and p. 69 'rsk wlgSy MLKyn M[LK'] 
are found in the Parthian period. Cf. G. A. 'Arsaces Vologeses King of Kings,' in R. Ghirsh- 
Koshelenko, "Parfyanskaya fortifikatsiya," SA no. man, Iran, Parthians and Sassanians (N.Y., 1962). 



The Parthians on the Plateau 241 

We have no Parthian literary remains from the Parthian period, but the existence 
of a large poetic or minstrel oral literature has been cogently proposed. 147 The 
Iranian national epic as preserved in the Shahname of Firdosi is primarily of eastern 
Iranian origin and incorporates Parthian heroes such as King Gotarzes in the tales 
which have survived. Yet the epic is not concerned with the wars between the 
Parthians and the Romans, but with older struggles of the rulers of the east and the 
struggle between Iran and Turan, the latter an uncertain people and place in the east, 
but more mythical than real, later identified with the land of the Turks. The traces of 
Parthian culture and society in later literature such as the Middle Persian text the 
Draxt I AsQrlg 'Assyrian (Babylonian) tree,' and the Yadegdr I Zareran 'Memorial of 
Zarir,' and the New Persian poem, Vis u Ramitn, are what one would expect, a heroic, 
chivalric society which could be called 'feudal' in a general sense. This is also the 
society and culture depicted in Firdosi's epic, and the heritage is clearly from Parthian 
times. This society and culture has little influence from Hellenism other than possible 
borrowings of stories or motifs, of Herakles to Rustam for example, while the Iranian 
character of the epic is paramount. Thus by the end of the Parthian period the Iranian 
revival had absorbed Hellenistic elements but existed in its own right in domains of 
art and culture, not to mention government, religion and society. The history of the 
Parthians, however, cannot be divorced from that of their powerful eastern 
neighbors, the Kushans, with their great king Kanishka, or from the history of the 
small states in western Iran and Babylonia where archaeology has revealed much of 
Parthian influences, and to these we must now turn. 

147 M. Boyce, The Parthian gisto and Iranian "Zariadres and Zarer," BSOAS, 17 (1955), esp. 
Minstrel Tradition," JRAS (1957), 10-45, and her 476-77. 



CHAPTER IX 
THE KUSHANS 

Literature: Literary sources on the Kushans are almost non-existent, other than brief notices of 'Bactrians' 
in Classical sources, where the word 'Kushan' does not appear, and in Armenian and Chinese sources, in 
both of which the identifications of Kushans are highly conjectural.* Indian sources merely mention them 
in the sub-continent, and we are left with numismatics and a few inscriptions to aid us to reconstruct the 
history of the most important ancient dynasty in the eastern Iranians world down to the coming of Islam. 
The material remains of the Kushans, however, are numerous, especially with the famous and prolific 
'Gandharan' Buddhist art which influenced eastern as well as Central Asia. The secondary literature on the 
Kushans, however, is overwhelming. There have been several international symposia concentrating on 
the date of Kanishka, but general conferences on the Kushans have proliferated, the largest having been 
one in Dushanbe in 1968 which resulted in two volumes, ed. by B. Gafurov Tsentralnaya Aziya v 
Kushanskuyu Epokhu (Moscow, 1974-75), which summed up knowledge about the Kushans until that 
time. 1 A conference in Kabul in November 1978 and another in Delhi in January 1980 (limited to the 
Kushans in Mathura and India), indicate the general expanded interest in the eastern counterpart of the 
Roman Empire. Fussman's article in J A has been noted in the chapter on the Sakas, and his assessment of 
the high value of the general work by B. Ya. Staviskii, Kushanskaya Baktriya (Moscow, 1977) has been 
confirmed by others. Most late excavations in Afghanistan, Tajikistan and Pakistan uncover objects from 
the Kushan period, and they reveal the wealth and prosperity of the Kushan Empire, much more than 
what was found under their predecessors or successirs. For French excavations in Afghanistan, see the 
various numbers of the MDAFA and for the Soviet excavations in northern Afghanistan (southern 
Bactria) see I. T. Kruglikova, Drevnyaya Baktriya, 2 vols. (Moscow, 1976 and 1979). 2 On the site of 
Bamiyan see Z. Tarzi, L 'architecture et le decor ruprestre desgrottes de Bamiyan, 2 vols. (Paris, 1977). We may 
say that the Kushan sites of importance are: Begram Dilberdzhin and Surkh Kotal in Afghanistan, and 
Dalverzin Tepe, north of the Oxus, and on the Indian sub-continent the twin sites of Charsada, and 
Shaikhan-Dheri, while Taxila and Mathura should not be omitted, although finds from other periods 
predominated in the last two. 3 The journals Afghan Studies, published in London, Afghanistan in Kabul 
and Aryana (in Pashto and Persian) in Kabul keep the student abreast of work in pre-Islamic culture and 
civilization, while the Journals EW, Afghanistan Journal (Graz) and Journal of Central Asia (Islamabad) 

*The great inscription at Surkh Kotal, dating Dilberdzhin, 1 (Moscow, 1974) and 2 (Moscow, 

from the time of Huvishka, has not been discussed 1977) from the 1970-72 and 1973 excavations 

in this chapter, since it is reserved for an appendix. respectively. 

1 On the Kanishka congresses, see papers in the 3 Bibliography on Begram is found in the 
JRAS for 1912-14 and A. L. Basham, ed., Papers volumes of MDAFA; on Surkh Kotal see D. 
on the Date ofKaniska (Leiden, 1968), 478 pp. On Schlumberger, "Le temple de Surkh Kotal," I-IV, 
one Kushan conference in Kabul in 1970 see F. JA (1952, 1954, 1955,and 1964), as well as reports 
Ayubi, Kushan Culture and History (Kabul, 1971), in CRAI, and "The Excavations at Surkh Kotal," 
180 pp. For Soviet contributions to the field see Proceedings of the British Academy, 47 (London, 
B. Ya. Staviskii, et al, ed., Sovetskaya Arkheologiya 1961), 77-95. The site of Dilberdzhin Tepe has 
Srednei Azii i Kushanskaya Problema, 2 vols. been mentioned in note 2; for Dalverzin see G. 
(Moscow, 1968), an annotated bibliography. For Pugachenkova, Les tresors de Dalverzine Ttpi 
more bibliographies see I. Stwodah and A. Z. (Leningrad, 1978). On Charsada the excavation 
Modarrissi, Kushans, annotated bibliography, 2 report (with the same title as the name of the site) 
vols. (Kabul, 1978), and B. N. Puri, KusSna has been published by M. Wheeler (Oxford, 
Bibliography (Calcutta, 1977). For a bibliography 1 962), and "Shaikhan Dheri Excavations," by A. H. 
of Soviet works from the Dushanbe congress to Dani, Ancient Pakistan, 2 (Peshawar, 1966), 17— 
1977, see B. A. Litvinskii, "Problemy Istorii i 214. Taxila has been mentioned in chapter seven. 
Istorii Kultury Drevnei Srednei Azii v Svete For Mathura see K. L. Janert, ed., Mathura 
noveishikh rabot Sovetskikh Uchenykh," VDI, 4 Inscriptions by H. LUders (GSttingen, 1961), for 
(1978), 73-92. further references. 

2 On earlier work here see I. T. Kruglikova, 



250 Chapter IX 

have articles on ancient Afghanistan from time to time. F. R. Allchin and N. Hammond, eds.. The 
Archaeology of Afghanistan [ch. 6, n. 43], have a large bibliography on the Kushan period. For Tajikistan 
and Uzbekistan (northern Bactria), and to a lesser extent Turkmenistan and Kirghizia, the journals of the 
respective academies of science and universities should be consulted, as well as general works, such as the 
yearly survey of archaeology in the USSR, published by the Institute of Archaeology in Moscow, entitled 
Arkheologicheskie Otkrytiya, ed. by B. A. Rybakov. Individual works are mentioned below in footnotes on 
special problems. 



ORIGINS 

It is now generally accepted that the Kushans were one of the tribes known to the 
Chinese sources as YUeh-chih who moved from their original homeland on the 
western borders of Kansu province of China under pressure of their northern 
neighbors to the west about 175 B.C. There is no reason to doubt this information 
from the Shih-chi, and the annals of the former and later Han dynasty, although there 
are points of difference between them. 4 The Yueh-chih invaded West Turkestan and 
apparently were among the tribes which brought an end to the Greco-Bactrian rule 
in that area and northern Afghanistan about 1 28 b.c. as mentioned above. Nomadic 
invasions of the western lands from Central Asia always carried many peoples along 
the way, such that frequently a billiard-ball effect was produced with one group 
impinging on another to send them moving ahead. The Sakas probably were the first 
nomads to invade the Greco-Bactrian domains but, according to Strabo (XI, 51 1), the 
Asii, Pasiani, Tokhari and Sakarauli took Bactria from the Greeks. Undoubtedly 
different groups took part in the movement, including Tokhari, who gave their name 
to the mountainous area of eastern Bactria, later called Tokharistan. 5 Whether some 
of the nomads spoke 'Tokharian,' the centum Indo-European language, is uncertain 
but probable, but the local Iranian tongues were adopted by the invaders, although 
debased Greek continued to be used on coins. The end of the Greco-Bactrians had 
brought a loss in Classical sources about events in Central Asia, and Chinese sources, 
unfortunately, are of little aid in reconstructing the past in this area. For the rise of the 
Kushans, however, a notice in the Hou (Later) Han Shu is important. It says that: 
"Formerly, when the YUeh-chih had been routed by the Hsiung-nu, they moved to 
Ta-hsia and divided the country into five hsi hou (yabghu): Hsin-mi, Shuang-mi, 
Kuei-shuang, Hsi (or Pa)-tun, and Tu-mi. More than a hundred years passed, the 
yabghu of Kuei-shuang, (called) Ch'iu-chiu-ch'deh, attacked and destroyed (the 
other) four yabghu, set himself up as king. The kingdom was called Kuei-shuang. The 
king invaded An-hsi, took the country of Kao-fu. He also destroyed P u-ta and Chi- 
pin, and completely subjugated them. Ch 'iu-chiu-ch 'iieh died at the age of more than 

4 On recent translations of the Chinese sources, dispute over the language and identity of the 
see E. Ztlrcher, "The YUeh-chih and Kaniska in the YUeh-chih, whether they were Iranian or Tokhar- 
Chinese Sources," in Basham, op. cit.[n. 1], 358-71. ian speakers, similar to the Wu-sun, Ta-yUan, 
See also the translations by K. Enoki, in Narain, K ang-chii and others, ace. to Pulleyblank, "Chi- 
The Indo-Creeks [supra, ch. 7, n. 2], 129-31. Cf. nese and Indo-Europeans," [ch. 7, n. 42], esp. 27- 
Pulleyblank, "The Wu-sun and Sakas" (ch. 7, n. 36. The latter identifies the YUeh-chih with the 
38], 1 54-60. VctTtoi of Ptolemy and Ta-yUan with Tokharians. 

5 The controversies over the identifications and H. W. Bailey in many articles supports the Iranian 
locations of western place-names in Chinese identity of the YUeh-chih. 

sources cannot concern us here, as well as the 



The Kushans 251 

eighty. Yen-kao-chen succeeded him as king. He in turn destroyed T 'lln-chu (India) 
and placed a general to supervise and govern there. Since that time the Ytleh-chih 
have become most rich and prosperous. (People of) many countries speak of the king 
of Kuei-shuang, but in China they are called Ta Ytleh-chih, according to their old 
name." 6 The interpretation of this passage generally has been that about the end of the 
millennium a yabghu or chief of one of the five tribes or divisions of territory of the 
new domain of the Yueh-chih, called Ch 'iu-chiu-ch 'tleh, who has been identified as 
Kujula Kadphises, conquered the others and established the Kuei-shuang (Kushan) 
kingdom. Much controversy has existed over the meaning and location of the people 
ruled by the yabghus, a Central Asian title, whether they were only tribes or also 
settled folk. 7 If we accept the statement of the text that over a hundred years after the 
conquest of Ta-hsia, usually identified as Bactria, the events described happened, then 
it is most likely by that time the people were settled and mixed, and the notice of the 
Ch Ve« (Earlier) Han Shu that the five yabghus belonged to the Ta Yueh-chih, as their 
subjects, probably means that the Ta Ytleh-chih included both the tribes with their 
leaders, and the settled areas over which they ruled. 8 Many attempts have been made 
to identify the Chinese characters with place names in Central Asia or Afghanistan, 
but it would seem we have in the text a mixture of place and tribal names, with great 
difficulties in identification. More likely for identifications than tribes or regions 
would be an identification of the capital towns of the five yabghus, but even they 
present great difficulties, since texts in other languages, except Ptolemy's Geography, 
give us no clues to city names in this region, and Ptolemy unfortunately does not 
provide parallels to the Chinese place names. 

From the meager reports of Chinese sources we must turn to the backbone of 
Kushan studies, as it was in the case of the Sakas, numismatics. Here, the sequence of 
Kushan coins has been established on the basis of styles and quality of the coins more 
than on legends. After the fall of the Greco-Bactrian kingdom, 'barbarous' copies of 
the coinage of Eucratides and especially of Heliocles were struck in Bactria, but 
whether north or south of the Oxus River is unknown. 9 Sometime in the first 
century B.C., or possibly at the turn of the millennium, probably it was one of 
the Yueh-chih chieftains who struck tetradrachms and obols with a corrupt Greek 

6 ZUrcher, op. cit. [n. 4], 367; Narairi, op. cit. substitutes the yabghu (meaning either the leader or 
[supra, ch. 7, n. 2], 131. Differences in translation his clan, or the land he ruled) of Kao-fu for Tu-mi. 
are minor but interpretations of proper names and The former has been identified as Kabul and the 
their significance vary widely. latter as Termez (ancient Tarmita). The ancient 

7 Cf. the article by P. Daffina, "The Return of transcriptions of the Chinese characters following 
the Dead," EW 22 (1 972) 87-91 , attacking B. N. Karlgren, are given by ZUrcher, op. cit. [n. 4], 388- 
Mukherjee, "The Ta-hsia and the Problem Con- 90. 

cerning the Advent of Nomadic Peoples in Greek 9 Cf. A. M. Mandelshtam, "K predkushans- 

Bactria" in EW, 29 (1969), 395^00, both with komu chekan Baton," EV, 17 (1965), 85-91. The 

further references. The word yabghu is considered special monogram on the copies of Heliocles coins 

Tokharian, comparing A dialect yapoy, B dialect may have been a mark or tamgha of a nomadic 

ype 'country' by Pulleyblank, op. cit. [ch. 7, n. 42], chieftain, since it is not found on any Greco- 

28, and as Iranian by H. W. Bailey, in "Languages Bactrian coin. It is hardly possible to identify or 

of the Saka," HO, Iranistik (1968), 136, comparing locate in place or time the enigmatic coins with 

Parthian ymg 'leader' and the root yam- 'to lead.' debased Greek legends and names such as Hyr- 

Others erroneously consider the word as Turkic in kodes ; cf. Tarn, Greeks [ch. 1 , n. 23], 301 , n. 1 , and 

origin. 305, n. 2, for further references. 

8 The former Han Shu in the list of tribes 



Z5Z Chapter IX 

legend to be reconstructed as TYPANNOYNTOZ HPAOY ZANAB{OY) 
KOPPANOY (with P for the sound -s-). 10 The title of Heraus or Heraeus is less than 
a king, of course, but exactly what it meant is uncertain. The name was much 
disputed because of the variations in corrupt legends, and the name is not found 
elsewhere unless it is simply a variation of Greek Heras. The third word, too, is 
enigmatic, but since the title was given in the first word, we may suggest that Sanab, 
as in the name Sanabares, is a family name followed by the tribal name of 'Kushan.' In 
any case, the person seems to have been a Kushan chieftain, and he has been identified 
as the person of whom a head and statue in clay have been found at the site of 
Khalchayan on the Surkhan Darya in Tajikistan. 1 ' Just where and how long Heraus 
ruled is unknown, but the next Kushan ruler of whom we have coins was Kujula 
Kadphises. His coins are varied in their origins, for he copied coins of Hermaeus, or 
pseudo-Hermaeus, Gondophares and others, indicating that he followed local usage in 
his conquests. This ruler is identified as the one called Ch'iu-chiu-ch'lieh (Archaic 
Chinese reconstruction: k'yug-dz^idg-k'yak), who united the yabghus of the Yueh- 
chih, invaded An-hsi (Parthia) and took Kabul from them. 12 This would refer to the 
Indo-Parthians, about the time of Gondophares. Most scholars agree that Kujula made 
raids into the lowlands of India and perhaps established rule for some years in various 
localities, but apparently he abandoned some of his conquests. Since he is said to have 
lived to an age of eighty years, presumably he had a long reign. Since few of his coins 
are found north of the Oxus River, and many of his coin-types have Kharoshthi 
legends as well as Greek, we may suppose that he ruled primarily in southern Bactria, 
in the Hindukush region and in Gandhara, on the northwest plains of the sub- 
continent. Kujula probably lived at the same time as Gondophares but how far their 
reigns and territories overlapped are unknown. The existence of copper coins of 
Kujula copied from Roman coins of Augustus or Tiberius, however, suggests a date 
for Kujula's reign in the first century of our era. 1 3 Kujula, however, does not seem to 
have been more than a conqueror seeking booty, perhaps establishing his rule and 
then withdrawing, or possibly again returning, and copying the coins of his 
predecessors in every area which he overran. Inscriptions in Kharoshthi are 
unfortunately of little help in historical identifications; for example, one from Takht- 
i Bahi on a stone in the Lahore Museum mentions a prince (Prakrit erjhutia) called 
Kapa, who has been identified with Kujula Kadphises, but this is most uncertain, as is 
the date. 14 

10 Cf. E. A. Davidovich, Klady drevnykh i perhaps meaning 'younger' or 'shorter,' and Kad- 
srednevekovykh monet Tadzhikistana (Moscow, phises as *gaSa pisa 'he who is adorned with the 
1979), 17-35, with references. See also R. B. throwing club,' see W. Eilers, "Die Namen der 
Whitehead, "Notes on the Indo-Greeks," NC, fifth Kuschan-Konige," Afljali, Wijesekera Felicitation 
series, 20 (1940), 120-22. Apparently no copper Volume (Colombo, Sri Lanka, 1970), 119, 126-27. 
coins of this ruler have been found. H. W. Bailey in "Irano-Indica III," BSOAS, 13 

11 See G. Pugachenkova, Skulptura Khalchayana (1950), 396-97, interprets Kadphises as Old Ir. 
(Moscow, 1971), 57 and plates 61-64, also her "K *kata-paisa- 'of honored form.' 

ikonografii Geraya," VDI, 1 (1965), 127-36. 13 D. W. MacDowall, "Numismatic Evidence 

Heraus copied the corrupt tetradrachms of Eucra- for the Date of Kaniska," in Basham, op. cit. [n. 1], 

tides, as others copied the coinage of Hehocles. 1 44—45. 

1 2 For the reconstructed Chinese, see ZUrcher, 14 S. Konow, Kharoshthi Inscriptions, Corpus 
op. cit. [n. 4], 389. On the name Kujula, Ko£o\o, Inscriptionum Indicarum (Calcutta, 1929), 62. 



The Kushans 253 

Inscriptions may have dates but the problem of the eras in which they are dated has 
plagued scholars for many years, and while relative chronologies can be established 
for the Kushan rulers, absolute dating eludes us, principally because dates found in the 
Indian sub-continent cannot be attached to chronologies in the west except by 
inferences and 'common sense' deductions. The tendency to seek the beginning of an 
era in an important historical event, or in the crowning of a king, is logical but quite 
uncertain, since the beginning of the Seleucid era had no such important beginning, at 
least in the eyes of modern scholars. Perhaps the most generally accepted beginning 
date of an era is one called the 'Saka' era, which is supposed to have started in the year 
78 of our era. When one closely examines how the date 78 was determined, however, 
one's confidence in the sources for the date, the Buddhist texts, may be somewhat 
shaken, since the sources are late and are based on a theory of the turning of the wheel 
of Dharma according to a tradition of a prediction by the Buddha forty years before 
his death, that five hundred years later the Buddhist law would come to an end. By 
interpolating dates and Buddhist information Eggermont decided that the 
Theravadin Buddhists of Ceylon used an era beginning in 483 B.C. whereas the 
Sarvastavadin school of northern India used one beginning in 383 B.C., and the date of 
a council convened by Kanishka, which began a new era after 500 years, took place in 
a.d. 78. ' 5 Unfortunately, none of the inscriptions informs us which era was used in it 
for dating, so even if the year 78 were correct, it would not solve most chronological 
problems of length of rule, because other eras exist in India from this time which 
cannot be ignored. The Vikrama Samvat, or era of Vikrama, may have been 
established by a ruler called Vikramaditya who is supposed to have defeated some 
Sakas in East Rajputana about 57 or 58 B.C., although some scholars believe this to be a 
late tradition and that originally this era was started by Azes, a Saka ruler. 16 
Paleography can aid in dating inscriptions, but the creation of new eras to justify an 
interpretation of obscure words in an inscription has caused great problems in trying 
to fix the chronology, and the truth is that we cannot prove the date of 
commencement of any era. An era older than the Vikrama- Azes era is implied by 
dated Kharoshthi inscriptions which cannot be dated by the Vikrama era, and this 
older era has been called the Yavana era by Dobbins (note 16), the Indo-Bactrian era 
by Bivar, or the 'Old Saka' era by others, and the date of its beginning has been set at 
170 or 155 B.C. as well as other dates plus or minus. 17 A third system of time 
reckoning has been attributed to Kanishka, since all the inscriptions, dated from the 

1 5 P. H. L. Eggermont, "The Origin of the Saka bered that references to the Vikrama and Saka eras 
Era," in Akten des 24. Internationalen Orientalislen- are much later. 

Kongresses, ed. by H. Franke (Wiesbaden, 1959), 17 A. D. H. Bivar in G. Hambly, Central Asia 

540-43, and his "Kaniska, die Saka-Aera und die (London, 1969); 46, and W. E. van Wijk, "On 

KharosthT-Inschriften," ZDMG, 113 (1964), 559- dates in the Kanishka Era," Acta Orientalia, 5 

65. But see other evidence from al-Biruni in n. 17. (Leiden, 1926), 168-70; Mukherjee, op. cit. [n. 16], 

1 6 Cf. B. N. Mukherjee, Central and South Asian 32, and 44^15, for further references. BTrunT in his 
Documents on the Old Saka Era (Varanasi, 1 973) ; book on India (ed. by E. Sachau, 1 886, vol. 2, p. 6) 
27-31 , and K. W. Dobbins, "Eras of Gandhara," says the year 400 of the Yazdagird era equals 1088 
JOSA, 7 (1970), 257, and n. 24 for further of the Vikrama era and 953 of the Saka era, whence 
references. Mukherjee's argumentation for a new we arrive at 5-7 B.C. and A.D. 78 resp. for the two 
era, the Yavana era, beginning c. 170 B.C. (p. 32) is eras. 

ingenious but unconvincing. It must be remem- 



254 Chapter IX 

year 2 through 98, are in the Kushan period, mentioning rulers from Kanishka 
through Vasudeva. 18 Other eras may have existed in Central Asia, Afghanistan and 
northwest India in the pre-Islamic period, but for the history of the Kushans the three 
eras mentioned above are of importance. Dating according to the regnal year of a 
king, a widespread practice in this part of the world, also existed, but in the east we 
have relative and not absolute chronologies, so many problems still exist. It appears 
that several eras were in use for time reckoning in the east in one and the same region, 
just as the Seleucid and Parthian eras were used in the western part of the Iranian area, 
and possibly Kanishka, who was an innovator in many realms, was also the founder of 
a Kushan era. To return to the date 78 of our era, the beginning of the so-called 'Saka' 
era, many scholars have identified this date as the same year as the beginning of the era 
of Kanishka, while others dispute it. The first mention of the Saka era as such, 
however, is in an inscription southeast of Bombay (Badami) which is dated in the year 
500 'from the inauguration of the Saka king.' When it began is not stated. The dates of 
Kanishka have been the subject of many writings, including two international 
symposia and the general consensus now is that the beginning of his reign must be not 
earlier than 78 B.C. and not later than a.d. 150, although the latter date is much 
disputed. Attempts to place the beginning of Kanishka's rule in A.D. 225 or 278 have 
not received widespread support. 19 It has thus proved impossible to fix the beginning 
year of Kanishka's reign, but the turn of the first century of our era seems to be the 
period in which Kanishka lived, even though his dates of rule cannot be determined. 
The inscriptions presumably dated in the era of Kanishka are from the year 2 through 
23 in Kanishka's name, year 24 in the name of Vasishka, and years 28 through 60 in the 
name of Huvishka, and 64 through 98 in the name of Vasudeva, thus spanning a 
century. There are problems, including one inscription from the year 41 in the name 
of Kanishka, but that one may be dated in a different era. The eras mentioned above 
are those current in this general time period. 20 

18 For a list of the inscriptions see J. Rosenfield, Kharoshthi inscriptions, a 'Yavana' era of 155 B.C., 

The Dynastic An of the Kushans (Berkeley, Calif, a Pahlava (or Indo-Parthian) era of 88 B.C. and a 

1967), 264-73. This work is a good reference book Kushan (Kanishka) era of a.d. 103, but this 

on the Kushans. See also the three appendices to proliferation of eras creates new problems while 

Mukherjee, op. cil. [n. 16], for additional inscrip- hardly solving old ones. 

tions. See also B. Kumar, The Early Kusanas (Delhi, 20 The proposal of G. Fussman, "Documents 

1973), App. 1. epigraphiques Kouchans," BEFEO, 61 (1974), 41, 

For the date 225 see R. G6bl, "Zwei neue to date a series of inscriptions with numbers from 

Termini ftir ein zentrales Datum der Alten 270 to 399 as dates in the Arsacid era is an 

Geschichte Mittelasiens.dasjahr 1 des KusJnkOnigs ingenious attempt to explain their apparent time 

Kaniska," Anzeiger der phil.-hist. Klasse der Osier. period as Kushan, but he admits that inscriptions 

Akad. der Wiss. (Wien, 1964), 137-51, and for found as far east as Mathura would rather imply a 

278, E. V. Zeymal, "Nachalnaya data Kanishki-278 Greco-Bactrian era from the time of Diodotus. We 

g.n.e." in Gafurov, supra, Tsenlralnaya, 1,292-301, may, of course, have an Indian era. He gives 

with references, and the following article by V. G. arguments for the date a.d. 78 for Kanishka. The 

Lukonin in the same volume, 302-06. A conven- best analysis of chronological problems is by 

ient survey of the various theories of dating may Zeimal, op. cil. [n. 19], English summary, 136-61, 

be found in Kumar, op. cil. [n. 18], 58-77, and App. with systematic analyses of available data. His 

2, and in E. Zeimal, Kushanskaya Khronologiya conclusion, that Kanishka began his rule in a.d. 

(Moscow, 1968). A. K. Narain in his article on 278, however, causes many difficulties and is 

"The Date of Kanishka" in Basham, op. cil. [n. 1], unacceptable. 
237-39, proposes three different eras used in the 



The Kushans 255 

While the Kushan kingdom was being consolidated in the Bactrian region, to the 
north in the area of present Tashkent and the Jaxartes River another kingdom, called 
K'ang-chii (K'ang kid) in Chinese sources, was flourishing, but it is not possible to 
determine what kind of a state it was, a confederation of nomads or a more centralized 
state of sedentary people. 21 The people who organized this shadowy state may have 
been Sakas, Sarmatians, or even Wu-sun or Tokharian speakers, for hypotheses about 
the nature of the people are many and varied. Unfortunately, the Chinese sources give 
no information about the K'ang-chii other than the existence of a state to the north of 
the domains of the Yiieh-chih in western Turkestan, while archaeological excavations 
only attest the wealth and far-flung connections of the people who lived there, 
presumably throughout the period of Kushan hegemony to the south. 22 

To return to the Kushans, Kujula Kadphises was followed by a ruler who is only 
called Soter Megas 'great savior' on his coins but has the title 'king of kings' in Greek, 
higher than Kujula, and whose coins are abundant and found in excavations from 
northern Bactria to Mathura. Since his issues are varied as well as widely found, this 
ruler must have been important, and MacDowall has demonstrated from coin types 
and metrology that this ruler must follow Kujula Kadphises and precede Vima 
Kadphises while being contemporary, or shortly before, Pakores of the dynasty of 
Gondophares, since Pakores overstruck coins of Soter Megas. 23 Just who Soter Megas 
was has been debated, generally with three positions, first that he was Kujula, and his 
coinage represents a later phase of rule, second, he was Vima in the early years of his 
reign, or third, he was an independent Kushan sovereign somehow a link between the 
two Kadphises. 24 The position of the coins of Soter Megas between those of Kujula 
and Vima is now generally accepted by numismatists, although some would not go as 
far as MacDowall in calling the unknown ruler the Augustus of the Kushans, who 
earlier in his reign followed local coinage types and metrology but later issued a 
general coinage everywhere based on the Attic standard and using only Greek 
legends, and paving the way for Vima's reform of the currency. Some scholars have 
sought a way out of the dilemma by supposing the existence of co-kings, or 'senior' 
and 'junior' kings, as in the later Roman Empire, but again this is mere surmise 

21 On K'ang chtl see Litvinskii, "Das K'ang-chii- "Soter the Great - The Last of the Indo-Greek 

Sarmatische Farnah" [ch. 8, n. 9], 249-53. At times Kings," JNSl, 35 (1 973), 86-89, that the 'nameless 

K'ang-chii included Sogdiana, but an identification king was the last of the Greek kings with the 

of the two is improbable. Likewise identifications personal name Soter,' is unconvincing. On a Janus- 

with Kaxayai in Ptolemy (VI, 14, 11) or Kang- faced coin of Soter Megas and Vima see NC 

diz or Kangha of the Avesta help us little. (1892), 71, plate 15, no. 4. On the overstnkes of 

22 The profuse use of gold, including a corpse in Pakores on S.M., see A.SimonettainEW,8(1957), 

golden clothes 'chain or armor' and regalia found 49, plate 3, no. 1. Theories proposing that Soter 

at Issyk Kurgan near Alma Ata, probably date Megas was a Kushan vassal are unlikely, since his 

from the third century B.C. not the fifth, and may coins are so widespread, not to mention his 

represent the K'ang-chii culture or its predecessor. pretentious title. 

Cf. K. A. Akishev, Kurgan Issyk (Moscow, 1978), 24 MacDowall, op. tit. [n. 23], 252, also his 

61-78. The Pazaryk burials are also from this article on the same subject in JNSl, 30 (1968), 28- 

period. 48. Overstrikes of coins of Vima by Soter Megas 

23 D. MacDowall, "Implications for Kushan complicate the picture, ace. to A.Simonetta, EW, 9 

Chronology of the Numismatic Context of the (1958), 171, who prolongs the rule of Vima too 

Nameless King" in Gafurov, supra, Tsentralnaya, 1 , much. 
246-64. The suggestion of A. N. Oikonomides, 



256 Chapter IX 

without any evidence. 25 Mukherjee, who has written extensively on the Kushans, 
points out that on a Prakrit legend on one of the coins of Soter Megas we find 
maharajasa rajatirajasa tratarasa vamasa, the last word of which should be the name 
Vima, while the first three words would correspond to the Greek legend 'king of 
kings, savior.' 26 Whether Soter Megas should be identified as the young Vima on the 
basis of this uncertain Prakrit word is dubious, and judgment may be reserved on the 
identity or identities of the ruler(s) who issued coins with the legend 'Soter Megas, 
king of kings.' On the basis of find spots of those coins, however, one may conclude 
that Bactria, the Hindukush area and Gandhara were ruled by the early Kushan kings, 
while Arachosia and parts of India were ruled by Indo-Parthian or Saka kings, 
although Gandhara at times may have been divided between Kushans and others. 
With Vima Kadphises the great Kushans begin. 

THE GREAT KUSHANS 

Vima introduced a gold dinar to the coinage system of the Kushans which had been 
previously only copper, and he also introduced a new copper tetradrachm, both 
implying a change in Kushan history. 27 Both the Chinese sources and Indian 
inscriptions are ambiguous in that no name in them can be certainly identified as 
Vima Kadphises, and his identification as the son and successor of Kujula Kadphises is 
not without doubts, but whatever his relation to Kujula and to the person who issued 
the Soter Megas coins, he seems to have ruled in the last part of the first century A.D. 
His relation to rulers such as Zeionises or Jihonika (Prakrit: Jihuniasa) and others who 
struck coins may have been that of overlord to vassals, but again we are in the realm of 
conjecture. To repeat, we do not know whether the Kushans had a system of 'senior' 
and junior' kings like the 'Augustus' and 'Caesar' of the later Roman Empire, as 
suggested by some scholars to justify seeming conflicts in chronologies (relative, not 
absolute) of inscriptions and coin types, but satraps and perhaps vassal rulers surely 
existed. We may say little more than Vima's extensive possessions in India and 
Central Asia were passed on to Kanishka, the greatest and in a sense most enigmatic of 
the Kushan kings. 

Kanishka is generally credited with introducing a new family of Kushan rulers 
whose names all end in -ishka, for he is followed by Vasishka and Huvishka. This 
suffix of the superlative caused controversy, for some scholars proposed a Tokharian 
origin for it, while others supported an Iranian origin, although all agreed the general 
meaning of Kanishka was something like 'the youngest.' 28 Whether Kanishka began 
as a sub-king of Vima in Bactria or Tokharistan with the full name Maho-Kanishka 
(Candra-Kaniska in Khotanese), as Henning (op. cit., 87) suggests, is possible but 

25 Gobi, op. cit. [n. 19], 151, n. 2. The existence 27 The name Vima has been explained as 
of kfatrapas and maha ksatrapas lit. 'satrap' and 'great 'fearsome' by Eilers, op. cit. [n. 12], 120. On his 
satrap' in India, as well as the Indo-Parthian title monetary reform see MacDowall, "Weight Stan- 
'brother of the king,' may point to the possibility of dards of Kushana Coinages," JNSI, 22 (1960), 67- 
a double system of rule under the Kushans, but this 71 . 

is undocumented. 28 For the Tokharian origin supported by L. 

26 B. N. Mukherjee, The Kushana Genealogy Hertsenberg and V. V. Ivanov, see L. G. Gertsen- 
(Calcutta, 1967), 54. berg, "Kushanskii i Sakskii," in Gafurov, supra, 



The Kushans 251 

unproven. 29 In any case, Kanishka was remembered in Buddhist sources although in 
no recognizable historical context; nonetheless some scholars have attempted to 
derive a chronology from a late mention of Kanishka as the ruler who convened the 
third Buddhist council, which reportedly took place five hundred years after the death 
of the Buddha. 30 Obviously the greatest Kushan king made an impression on 
Buddhists, for we find traditions in Buddhist texts written in Khotanese-Saka, in 
Sogdian, and Chinese about his building of stupas (mounds with reliquary) and 
vihdras (cloisters), while Biruni mentions a King Kanik who built a vihctra in 
Purushavar (Peshawar), an indication of the fame of his constructions. 31 
Unfortunately, historical information cannot be found in the Buddhist legends and 
from the extant coins one may postulate a number of rulers with the name Kanishka, 
one of whom is mentioned in the poetical history of Kashmir of the twelfth century, 
Kalhana's Rajatarangini (Part 1, 168). As usual, coins provide the primary source for 
our knowledge of Kanishka. He continued the gold coinage of Vima, but he changed 
the obverses and the reverses on the coins. Vima had been portrayed on obverses 
seated, or as a bust, or riding a chariot or an elephant, while on all of his reverses the 
Indian deity Siva was pictured. Kanishka, however, inaugurated on his obverses a 
new type, a standing figure of the ruler in Central Asian costume with boots and with 
one hand extended over a small fire altar. This consistency on the obverse was 
matched by a proliferation of deities on reverses of his coinage, the majority of which 
are Iranian rather than Greek or Indian. This change in the stance of the ruler with the 
deities on the reverses, each with its name in Greek characters, supports the theory 
that Kanishka consciously promoted a pro-Iranian policy in his empire, or perhaps it 
meant an end to the past and a proclamation of a new policy of tolerance in religions, 
as well as a change from Greek legends on the early coins of Kanishka to Iranian 
legends on his later coins. It seems as though Kanishka, while copying Roman models 
in his coinage, and probably in art and other manifestations of culture, nonetheless 
wished to proclaim a new, imperial age in the east, possibly in imitation of the 
Achaemenids as suggested by Fussman, but surely with strong elements of the ancient 
Indo-Aryan culture as preserved by local Iranians or Iranicized nomads of the steppes, 



Tsentralnaya, 1 , 344—49 (French resume and candra 'moon' which may also involve a confusion 

reference to Ivanov), while the Iranian origin was between Iranian mao 'moon' of the Kushan coins 

upheld by H. W. Bailey, "Kanaiska,"JiMS (1942), and a word based on mau- meaning 'deliverer.' So 

16-47, also in Basham, op. cit. [n. 1], 37, by W. B. the assumption that Kanishka was called Msho 

Henning, "Surkh-kotal und Kaniska," ZDMC, must be regarded as nothing more than conjecture. 
115 (1965), 82-4, by Livshits and others. V. 30 See F. Wilhelm, "Kanika and Kaniska - 

Abaev, in "Contribution a l'histoire des mots," Asvaghosa and Mstrceta" in Basham, op. cit. [n. 1], 

Milanges linguistiques offerts a E. Benveniste, ed. by 337-40. 

M. Moinfar (Louvain, 1975), 7, showed the 31 For literary and archaeological references to 

widespread Indo-European usage of the suffix. See the buildings in Peshawar, see K. W. Dobbins, The 

also Eilers, op. cit. [n. 12], 112. Stupa and Vihara of Kanishka I (Calcutta, 1971), 

29 On the epithet 'candra' see Bailey, "Kanaiska," esp. 44-54. The fame and vast extent of Kanishka's 

op. cit., 16, and in his article "Saka Candarno and empire was known to the Chinese Buddhist 

Surkh Kotal Kara-Lrangga-," in Milanges linguist!- pilgrim of the seventh century HsOan Tsang; cf. T. 

ques offerts a E. Beneviste [n. 28], 33, where he Watters, trans 1., On Yuan Chwang's Travels, 1 

suggests that the Khotan-Saka word candarno (London, 1923), 270. 
'troop leader' was replaced by Indian Buddhist 



258 Chapter IX 

which the ancestors of the Kushans were. 32 This probably would appeal to eastern 
Iranians and Indians more than memories of the west Iranian empire of the 
Achaemenids, although old traditions and stories surely were preserved about them 
in the east. Hellenistic influences observable in the artistic productions were 
comparable to those in the Parthian state, but styles and techniques were borrowed 
from elsewhere in both cases, and in both cases the basic inspiration was Iranian. 
Under Kanishka, who ruled at least twenty-three years after establishing an era, to 
judge by dated inscriptions with his name in them, the Kushan Empire probably 
attained its apogee, although under Huvishka, who ruled from year 28 of the 
Kanishka era to at least 60, the empire was also far extended. To the north of Bactria, 
homeland of the Kushans, finds of coins may indicate the extent of the Kushan 
domination, since no written sources exist and archaeological finds are difficult to 
connect with the Kushans. A detailed study of early local Khwarazmian copper 
coinage, and overstrikes on Kushan coins found in excavations within the area of 
ancient Khwarazm, point to the conclusion that the Kushans neither ruled the area 
directly nor held it in a vassal relationship; rather, it would seem that the Kushan 
coins found in Khwarazmian sites were brought by trade. 33 The Zarafshan basin 
with its principal city Samarqand also seems to have not been under Kushan rule, 
although this is uncertain since not enough excavations have been carried out in the 
ancient city sites occupied by buildings today. Whereas Greco-Bactrian coins have 
been found in sufficient numbers to postulate their rule in ancient Sogdiana, Kushan 
coin finds are fewer, although from sites on the Oxus River more have been found. 34 
Only one coin of the 'great Kushans' has been found in Samarqand, and the extensive 
excavations of Panjikant east of Samarqand have yielded only several poor specimens 
of Kushan coins, which leads one to speculate that direct Kushan rule was absent, or 
possibly coins were little used and barter was more widely practiced than to the 
south. 35 Place names such as medieval Kushaniya, northwest of Samarqand, and 
others, may point to an occupation by the Kushans, but they tell us nothing about the 
area in the time of the 'great Kushan' rulers. 36 Sogdiana, as noted, may have been part 

32 In G. Fussman, "Documents epigraphiques 35 Cf. M. E. Masson, "Po povodu dalekogo 
Kouchans,"BEFEO,61 (1974),esp. 50 foil., and his proshlogo Samarkands," in G. A. Pugachenkova, 
"Le renouveau Iranien dans l'empire Kouchan," in Iz istorii iskusswa velikogo goroda (Tashkent, 1972), 
Le plateau iranien et I'asie centrale, colloques interna- 25, and 0. 1. Smirnova, Katalog monet s gorodishcha 
tionaux du Centre natl. de la recherche scientifique, no. Pendzhikent (Moscow, 1963), 59-60. 

567 (Pans, 1976), 313-22. His emphasis on the 36 M. E. Masson, "K Voprosu o severnykh 

anti-Greek character of the changes made by the granitsakh gosudarstva 'Velikikh Kushan," in 

Kushans, in my opinion, is exaggerated, since the Gafurov, supra, Tsentratnaya 2, 42—49 with an 

few, if any, Greeks in the empire in the first and English summary. Also the inscription of Shapur at 

second centuries of our era hardly can have had any the Ka 'bah of Zoroaster implies that the Kushans 

influence either politically or even socially, or did not rule Sogdiana in the middle of the 3rd 

directly cultural in the Kushan Empire. century. The Oxus River, however, probably was 

33 B. I. Vainberg, Monety Drevnego Khorezma ruled by the Kushans, at least as far as Khwarazm. 
(Moscow, 1977), 85-89. This study supersedes Cf. V. N. Pilipko, Topografiya nakhodok Ku- 
older studies which supported the theory that the shanskikh monet na poberezhe Srednei Amudari," 
Kushans directly ruled Khwarazm, e.g., by Tol- in Istoriya i Arkheologiya Srednei Azii,ed. by O. V. 
stov, op. cit. [ch. 8, n. 8], 180-84. For further Obelchenko (Ashkabad, 1978), 89-97. 
bibliography see Vainberg, 193. 

34 Cf. V. A. Shishkin, Varakhsha (Moscow, 
1963), 229, with other references. 



The Kushans 259 

of the state called K'ang-chU by Chinese sources, the center of which was to the north, 
but the information is simply not explicit and one can only conjecture. 

Chinese Turkestan, with more literary fragments than Sogdiana, is also an 
unknown area in regard to Kushan rule, although Kushan cultural influences, 
especially through Buddhism, were undoubtedly strong. Commerce was lively 
between east and west and from India to China through Sinkiang, at least from the 
time of the Greco-Bactrian state, although actual direct Greco-Bactrian rule in 
Chinese Turkestan is unlikely. 37 Likewise, for the Kushans no evidence for political 
rule there can be found, although religion joins trade and cultural contacts as 
testimony of closer contacts between the Kushan Empire and Chinese Turkestan. 
Bactrian loanwords in the Tokharian' language (A and B dialects), and the 
occurrence of the name of Kanishka in a local Tokharian B text suggest a possible rule 
of the Kushans in the oases northeast of Kashghar. 38 The use of the Kharoshthi script, 
the expansion of Buddhism to the east, probably in the Kushan period, and other 
similar indications have induced J. Brough to conclude that the cumulative effect was 
not the result of 'influences' but of the political domination of the Kushans over most 
of Sinkiang including areas far to the east of either Kucha in the north or Khotan in 
the south, including a state called Shan-shan near the western borders of China 
proper. 39 If true, it is not possible to do more than guess at a time period or the 
geographical extent of Kushan rule in Chinese Turkestan, not to mention the nature 
of such a rule, if it existed, but a satrap or more likely a vassal relationship of local 
rulers to the 'great Kushans' would be more consistent with what we know of 
political organization in that age rather than central control. Evidence for Kushan 
rule there, however, is only based on inference. 

Many scholars have argued that the Kushan Empire was primarily an Indian one, 
since the wealthiest part of it was in the sub-continent. No doubt, the majority of the 
population was in the lowlands of Gandhara, the Punjab and the Ganges basin, and 
Indian merchants, Buddhist monks and others from the sub-continent were found all 
over the Kushan Empire, but the rulers were not Indians, and the bases of rule were 
primarily Iranian, although as time progressed Indian elements grew in importance. 
The very name of the last 'great Kushan' ruler Vasudeva, indicates the adherence of 
this ruler to, or at least a sympathy for, a cult of Vishnu or Vasudeva. The 'great' 
Kushans ruled at Mathura but the extent of the empire to the east and south is 
unknown, although from the absence of coin finds one may suspect that Kushan rule 



37 J. Harmatta in "Sino-Indica," AAH, 12 more than the official use of an Indian Prakit as an 
(1 964), 11-15 thinks Greek rule in Sinkiang likely, administrative language there. 

but only evidence of trade or culture not of 39 J. Brough, "Comments on Third-Century 

political domination may be inferred. Shan-shan and the History of Buddhism," BSOAS, 

38 W. Winter, "Baktrische Lehnworter im 28 (1965), 597. Brough suggests that Kushan rule 
Tocharischen," Donum lndogermanicum, Feslchrifl probably was quite short, possibly at the end of the 
fur A. Scherer (Heidelberg, 1971), 220-21. The Later Han dynasty in the second or at the 
same title borne by the king of Kucha in Sinkiang beginning of the third century, but any Kushan 
as the 'great Kushan' kings, is a prime reason for the rule there is disputed by B. N. Mukherjee in 
argument that the Kushans ruled Chinese Turkes- "KharoshthT Documents of Shan-shan and the 
tan, but this by no means proves such rule, any KushJna Empire," AAH, 24 (1976), 93. 



260 Chapter IX 

did not extend into modern Bengal or into the Deccan. 40 It seems that after the 'great' 
Kushans, Kushan rule collapsed in areas to the east and south of Mathura if not in that 
important city itself, and Naga rulers succeeded them in the Mathura area, whereas 
elsewhere tribes, such as the Yaudheyas, became the heirs of the Kushans. 41 This does 
not mean that the Kushans vanished from central India, while some later rulers had 
high pretensions as heirs of the Kushans, but any rule exercised by later rulers in India 
claiming to be Kushans was small. The mention of a devaputra, king of kings, of the 
Kushans in the Allahabad Pillar inscription of the Gupta ruler Samudragupta in the 
late fourth century only suggests that a small ruler in central India used a pretentious 
title, and one may speculate that the term 'Kushan' became almost generic in India 
after the third century, much as Classical sources mention 'Bactrians' but never 
'Kushans.' 42 Just how long Kushan princes ruled in the western Punjab, Gandhara or 
elsewhere in the west is unclear, but in the far western regions of their domain the 
Sasanians became the heirs of the Kushans in the late third and fourth centuries. 

To return to Kanishka, what more can be said about him in addition to his 
prominent but unhistorical place in Buddhist tradition, his initiation of an era, and his 
change of language from Greek to Bactrian in Greek letters, which indicate his other 
changes in the Kushan Empire, which may earn him the appellation the 'Darius' of his 
time? Attempts to write a history of the early life of Kanishka, with his original home 
in Khotan, Chinese Turkestan, and his subsequent conquest of India, based on fanciful 
Tibetan and Chinese Buddhist works are unacceptable, and we have no evidence at all 
of his early life or status. 43 From inscriptions mentioning him we can say he ruled at 
least twenty-three years after establishing a new era of time reckoning, while 
suggestions that he was a viceroy of Vima Kadphises before succeeding Vima cannot 
be proved. The center of his empire was in Bactria, and we may assume that his 
domains extended north to the Hissar mountain range and up the Oxus River, to the 
west possibly to the area of Herat and, in India, as far east as Varanasi (Benares) where 
an inscription in the name of Kanishka has been found at Sarnath, and another in the 
south in Sanchi, perhaps including eastern Malwa, though no evidence exists for rule 
in the Deccan. 44 Even though the homeland of Kanishka was in Bactria, the Indian 
part of the empire provided the wealth and power for the development of an 
impressive civilization, the artistic results of which are known as Gandharan art, 
discussed below. 

40 Cf. B. N. Mukherjee, The Kushanas and the 4:5 Puri, op. cit. [n. 40], 35-37, with references. 
Deccan (Calcutta, 1968), esp. 100 and 109, and For a suggestion that Kanishka may have marched 
B. N. Puri, India under the Kushanas (Bombay, into Chinese Turkestan against the Han general 
1965), 69-76. Coin finds suggest that only the Pan Ch'ao in a.d. 90, see K. Enoki, "Hsieh, Fu- 
'great Kushans' ruled some parts of the Ganges Wang or Wang of the YOeh-shih. A Contribution 
valley and later Kushan rulers did not. to the Chronology of the Kushans," in Memoirs of 

41 B. N. Mukherjee, The Disintegration of the the Research Department of the Toyo Bunko, 26 
Kushiva Empire (Varanasi, 1976), 78, 86-88, and (Tokyo, 1968), 12. This too is based on inference. 
Puri, op. cit. [n. 40], 76. 44 Cf. Dobbins, op. cit. [n. 31 ], 7 and Mukherjee, 

42 For a discussion of the Allahabad inscription The Kushanas and the Deccan, 78, 108. Dobbins, 
with references see S. Chattopadhyaya, Early 24-32, also describes the 'reliquary of Kanishka' in 
History of North India (Calcutta, 1958), 147-64, detail, mentioned below. On Kanishka's territories 
esp. 160. For the use of the word 'Bactrian' for in southern and western India, see the discussion by 
'Kushan' see, inter alia, Scriptores Historiae Augustae, Puri, op. cit. [n. 40], 52-54. 

Aurelian, 33, 5; Hadrian 21, 14; Valerian, etc. 



The Kushans 261 

Kanishka ruled at least a quarter of a century, and two inscriptions, one in Surkh 
Kotal, the other on a reliquary from a stupa in Peshawar, mention him, and since both 
are probably from a later date they indicate at the least the esteem in which the name 
Kanishka was held after his time. If we follow dates and names on inscriptions, it 
seems that Kanishka was followed by Vasishka, who probably ruled from 24-28 in 
the relative chronology of the 'great Kushans,' but who left no coins in his name, 
which may indicate a subordinate status as ruler of only Indian parts of the empire 
without the right to issue coins. 45 We cannot determine whether the empire was torn 
by civil war after the death of Kanishka or whether Vasishka and Huvishka were sub- 
kings, or vassals of Kanishka, but in any case Huvishka (meaning 'the best' or 'the 
oldest') appears on inscriptions from 28 to 60, and his coinage is plentiful in gold and 
copper found over the entire empire. His rule of more than thirty years (possibly 
before 28 and after 60) indicates a stable and nourishing empire. There exists a most 
enigmatic Kharoshthi inscription from Ara, north of Taxila, dated in the year 41 of 
Kanishka son of Vajheshka, who not only holds the usual exalted titles of a great ruler, 
but also a strange word reconstructed as 'Kaisar,' possibly a copy of Roman 'Caesar'. 46 
Many explanations of the information in this inscription have been proposed, 
including the existence of a later era of reckoning for the Kushans, as well as a short- 
lived ruler called Kanishka who usurped power from Huvishka, but from this one 
inscription no deductions about the succession can be made. 

The last of the 'great Kushans' was Vasudeva, who from his name, associated with 
the Hindu god Vishnu, shows an increase in the Indianization of the dynasty. His 
inscriptions are dated from 67 to 98 of the 'great Kushan' era, although again he may 
have ruled earlier and later than these dates. The name Vasudeva occurs among later 
Kushan rulers, and it is sometimes difficult to know to which ruler certain coins with 
this name refers, and disagreement exists among scholars on the assignment of some of 
the coins to one or another Vasudeva. Archaeologists agree, however, that the end of 
the reign of Vasudeva I marked a period of decline in the sites of Bactria, homeland of 
the dynasty, for traces of ruin in the strata dating to the end of Vasudeva's reign may 
be found in a number of Kushan sites of northern Bactria. 47 Both from the 
archaeological evidence and the seeming end of the era of time reckoning founded by 
Kanishka, with no inscriptions dated after the year 99 of that era, we may conclude 
that the century of the 'great Kushans' came to an end. Many scholars have dated this 
decline with the rise of the Sasanians, which is a good surmise but which leaves us with 
a latitude of many years in which to try to date the 'great Kushans.' All we can say is 
that the second century of our era, when the Parthians were not faring well with 
the Romans, was the century of 'great Kushan' flourishing, with exact dates still 
uncertain. 



45 Vasishka, meaning 'most energetic' (accord- op. cit. [n. 18], 59-60, discusses various interpreta- 
ing to Bailey in Basham [n. 1 ], 38, and Eilers, op. cit. tions of the inscription, not one of them more 
[n. 12], 1 17), is mentioned in a Brahmi inscription convincing than another. 

from Mathura dated 24 as mahamja rajatirap 47 See G. Pugachenkova, "K stratigrafii novykh 

devdputra, not a subordinate title. monetnykh nakhodok iz Severnoi Baktrii," VDI, 

46 Konow, op. cit. [n. 14], 162-65. Rosenfield, 3 (1967), 88. 



Abl Chapter IX 

DISINTEGRATION OF THE EMPIRE 

Shapur, son of Ardashir, in his inscription on the Ka *bah of Zoroaster, says he held in 
his empire 'the land of the Kushans (line 4 : Kovorp\wv i&]vr), Parthian kwSnxkr, MP 
missing), until in front of (iws 'ifxirpoad^ev) Pashkibur, and up to Kash, Soghd and 
the mountains of Chach,' which seems to indicate that about the year 265, the 
Sasanians had in fact secured control of, if not annexed, the Kushan domains up to 
Pashkibur, Kash, Soghd and the mountains of Chach or Shash. 48 The problems of 
identifying these places are many, but everyone would agree that Bactria, the 
homeland of the Kushans, the Hindukush mountains, and probably Kabul and some 
of the lowlands at Jalalabad and farther, were under some sort of Sasanian control at 
this time. 49 From what little we know of Indian history, presumably the Kushans 
continued to rule in parts of the Punjab or Northwest province after the end of rule of 
the 'great Kushans,' but coin finds on which much of the history of this period are 
based are notoriously unreliable, especially after an era of great, imperial expansion 
when copying of imperial coinage in many areas would be expected. Whether we 
must postulate a Kanishka III, a Vasishka II, and a Vasudeva II and III among later 
Kushan rulers is conjectural, but the retreat of the Kushans from regions east of 
Mathura seems to have begun already in the reign of Huvishka. 50 Just as the 
archaeological evidence from northern Bactria implies a great decline in Kushan 
fortunes under Vasudeva I, so the decline of trade between the Roman and Kushan 
Empires may also be attributed to the end of the reign of Vasudeva. 51 The continued 
rule of a Kushan king, Vasudeva II, in India after the Sasanian conquest of the western 
part of the empire has been postulated by several scholars, but it is impossible to do 
more than surmise the existence of rulers who followed Kushan traditions in coinage 
and in art, such as statues. The problem of various coins with the name Vasudeva but 
in different styles points to a copying of Vasudeva coins by others after him, similar to 
the coins of Hermaeus, last of the Indo-Greeks. Some scholars have proposed two 
branches of rulers copying Vasudeva coins, one following those coins with a triratna 
'three jewels' symbol on them, which eventually were copied by the Sasanian 
governors of the western Kushan domains in the fourth century, while the second 
group of Vasudeva coins with Brahmi akfaras 'syllables' on them developed into coins 
of the Murundas and Guptas in India. G6bl has developed an elaborate scheme of late 
Kushan rulers comparable to the late Roman Empire with a senior Augustus and a 
junior Augustus, as well as a division of the empire into two parts according to the 
two types of coins, but it is all conjecture. 52 

48 The basic publication of the inscription in 50 See Mukherjee, Disintegration [n. 41], 40-42, 
three versions, Greek, Parthian and defective and Puri, op. cit. [n. 40], 75-76. On Vasishka II see 
Middle Persian is M. Sprengling, Third Century R. Gobi, "VSsiska II," Anzeiger der phil.-hist. Klasse 
Iran, Sapor and Kartir (C